Non-alcoholic fatty liver disease may not be a severe disease at presentation among Asian Indians

Transcription

1 PO Box 2345, Beijig , Chia World J Gastroeterol 2006 Jue 7; 12(21): World Joural of Gastroeterology ISSN wjg@wjget.com 2006 The WJG Press. All rights reserved. CLINICAL RESEARCH No-alcoholic fatty liver disease may ot be a severe disease at presetatio amog Asia Idias Kaushal Mada, Yogesh Batra, S Datta Gupta, Bal Chader, K D Aad Raja, M S Tewatia, S K Pada, S K Acharya Kaushal Mada, Yogesh Batra, S K Acharya, Departmet of Gastroeterology, All Idia Istitute of Medical Scieces, New Delhi, Idia S Datta Gupta, Bal Chader, K D Aad Raja, M S Tewatia, S K Pada, Departmet of Pathology, All Idia Istitute of Medical Scieces, New Delhi, Idia Correspodece to: S K Acharya, Professor ad Head of Departmet of Gastroeterology, All Idia Istitute of Medical Scieces, Asari Nagar, New Delhi , Idia. subratacharya20004@yahoo.com Telephoe: Fax: Received: Accepted: Abstract AIM: To evaluate the cliical ad biochemical profile of patiets with o alcoholic fatty liver disease (NAFLD) ad to assess their histological severity at presetatio. METHODS: Cosecutive patiets presetig to the liver cliic of All Idia Istitute of Medical Scieces (AIIMS) with raised trasamiases to at least 1.5 times upper limit of ormal, ad histologically cofirmed o-alcoholic fatty liver disease were icluded. Patiets who had sigificat alcohol itake or positive markers of other liver diseases or who were takig drugs kow to produce fatty liver were excluded. The cliical, biochemical ad histological profile of this group was studied. RESULTS: Fifty-oe patiets with NAFLD formed the study populatio. Their media age ad BMI were 34(17-58) years ad 26.7( ) kg/m 2 respectively ad 46 (90.1%) were males. The majority of the patiets had mild iflammatio, either grade 1 [32 (63%)] or grade 2 [16 (31%)] ad oly 3 (6%) patiets had severe (grade 3) iflammatio. Twety-three (45%), 19 (37%), 8(16%) ad 1(2%) patiet had stage 0, 1, 2 ad 3 fibrosis respectively o idex biopsy ad oe had cirrhosis. O uivariate aalysis, triglyceride levels more tha 150 mg % (OR = 7.1; 95% CI: , P = 0.002) ad AST/ ALT ratio > 1 (OR = 14.3; 95% CI: , P = 0.008) were associated with high grades of iflammatio ad oe was associated with advaced fibrosis. O multivariate logistic regressio aalysis, hypertriglyceridemia >150 mg% was the oly factor idepedetly associated with presece of high grade of iflammatio (OR = 1.6; 95% CI: , P = 0.02), while oe was associated with advaced fibrosis. Triglyceride levels correlated positively with iflammatory grade (r = 0.412; P = 0.003). CONCLUSION: NAFLD i North Idia patiets is a disease of youg over-weight males, most of whom are isuli resistat ad they ted to have a mild histological disease at presetatio The WJG Press. All rights reserved. Key words: Iflammatio; Fibrosis; Triglycerides; No alcoholic steatohepatitis Mada K, Batra Y, Gupta SD, Chader B, Aad Raja KD, Tewatia MS, Pada SK, Acharya SK. No-alcoholic fatty liver disease may ot be a severe disease at presetatio amog Asia Idias. World J Gastroeterol 2006; 12(21): INTRODUCTION With the icreasig awareess ad early recogitio of asymptomatic patiets with raised trasamiases ad the ogoig epidemic of the metabolic sydrome, o alcoholic fatty liver disease (NAFLD) has become a commo cause of referral to hepatology cliics [1]. NAFLD ca be divided ito 4 histological types (types 1 to 4) [2]. Studies from the west have demostrated that NAFLD, especially NAFLD type 3 ad 4 ca progress over a variable period, i 25%-40% patiets, to cirrhosis ad ca cotribute to liver-related mortality [2,3]. Recetly there have bee reports likig the developmet of hepatocellular carcioma to o alcoholic steatohepatitis (NASH) [4,5]. Studies from Europe ad Uited States have demostrated that up to 40% patiets with NAFLD have advaced fibrosis ad a proportio eve have cirrhosis o the idex biopsy [6,7]. The prevalece of fatty liver i the geeral populatio of Idia has bee show to be as high as 24% [8], which is similar to that reported from some of the wester coutries, where it parallels the prevalece of obesity [9,10]. Despite the high prevalece of NAFLD i Idia ad rapidly risig icidece of type 2 diabetes mellitus i this coutry, data o hepatic morphology, to idicate the severity of NAFLD is limited [11-13]. The above-metioed 3 studies reported demographic ad biochemical profile of patiets with NAFLD. Histological parameters were studied i oly a small umber of patiets. By ow it is clear that histological type 3 ad 4 NAFLD are

2 Mada K et al. Histological severity of NAFLD i Asia Idias 3401 associated with progressive liver disease ad therefore eeds therapeutic itervetios. The preset prospective study was desiged to assess the magitude of advaced liver damage ad fibrosis amog cosecutive patiets with biopsy prove NAFLD. MATERIALS AND METHODS Study desig It is a cross sectioal study to evaluate the cliical, biochemical ad histological profile of patiets with NAFLD. The study was approved by the ethics committee of the All Idia Istitute of Medical Scieces(AIIMS), New Delhi. Patiets Iclusio criteria: Cosecutive patiets presetig to the liver cliic of AIIMS, with raised trasamiases to at least 1.5 times upper limit of ormal, ad histologically cofirmed NAFLD, were icluded. Exclusio criteria: Patiets i whom alcohol itake exceeded 20 grams per day (history of alcohol itake was take separately from the patiets ad the closest relatives), patiets with positive markers for other liver diseases (hepatitis viruses A through E, autoimmue, Wilso s disease, alpha 1 ati-trypsi deficiecy, hemochromatosis) ad patiets who were o medicatios kow to iduce fatty liver such as methotrexate, estroges, amiodaroe ad tamoxife were excluded. Methods All patiets were evaluated with a thorough history ad examiatio especially to exclude itake of sigificat amouts of alcohol ad to exclude itake of potetially hepatotoxic drugs. Athropometric measuremets were take at the iitial visit. These icluded the weight, height, waist ad hip circumfereces. Body mass idex (BMI) ad waist hip ratios were calculated as weight/height 2. After a overight fast, 10 ml of blood was collected for a complete blood cout ad biochemical ivestigatios icludig a liver fuctio profile, lipid profile, fastig serum isuli ad fastig blood glucose. Hepatitis virus serologies icluded hepatitis B surface atige (HBsAg), total atibodies to hepatitis B core atige (total ati-hbc) ad atibodies to hepatitis C virus (ati- HCV). Ati-HCV was doe usig a sesitive commercial ELISA (Xcyto, Bagalore, Idia) [12] ad the remaiig viral serologies were doe usig commercial ELISA kits (Orgao Tekika, Boxtel, The Netherlads). Serum ferriti ad 24-h uriary copper estimatio were doe usig covetioal biochemical techiques [13,14]. Alpha oe ati-trypsi (α1at) pheotypig was doe usig isoelectric focusig (PHAST system; Pharmacia Biotech, Uppsala, Swede) [15]. The ati-uclear atibodies (ANA), ati- smooth muscle atibodies (ASMA) ad ati-liver kidey ad microsomal atibodies (ati-lkm1) were carried out usig the immuofluorescece techique. Hepatitis C virus RNA was detected usig the polymerase chai reactio method as stadardized at our laboratory [15]. Liver biopsy was doe after a iformed coset i all patiets suspected to have NAFLD, usig a 18-gauge Meghii s aspiratio eedle. Each biopsy specime, after beig fixed i 10% formali, had routie hematoxyli-eosi stais ad special stais such as reticuli, Masso s trichrome, Perl s iro stai ad Orcei stai for copper ad periodic acid Schiff after digestio with diastase. Immuoperoxidase staiig to detect HBsAg ad hepatitis B core atige (HBcAg) was carried out i all patiets to exclude the presece of chroic viral hepatitis. The classificatio give by Brut et al was used to grade ad stage NASH [16]. A ultrasoography sca of the upper abdome was doe i all patiets i order to examie presece of fatty liver ad to exclude ay obstructive biliary pathology or the presece of ay hepatic space occupyig lesio. Fastig serum isuli values were used alog with simultaeous fastig blood glucose measuremets to calculate HOMA-IR (homeostasis model assessmet) accordig to the formula give below. HOMA-IR = (fastig serum isuli (μiu/ml) fastig plasma glucose (mg%))/( ) Value of HOMA-IR more tha 1.64 implied the presece of abormally high isuli resistace [17]. Defiitios Diabetes was regarded to be preset if the fastig veous plasma glucose was more tha 126 mg% [18]. Patiets with BMI of more tha 23 were see to be overweight ad those with a BMI of >25 were labeled as obese (Asia stadards) [19]. Hypertriglyceridemia was defied as a fastig serum triglyceride level of more tha 150 mg%. Patiets who had at least three of the followig five compoets: hyperglycemia (fastig blood sugar >110 mg% or kow type 2 diabetes mellitus, cetral obesity (as defied by a higher waist-hip ratio), hypertesio, hypertriglyceridemia (serum triglyceride > 150 mg%), ad low HDL cholesterol levels (< 50 for wome ad < 40 for me), were labeled to have the metabolic sydrome. Sice we did ot have waist hip ratio i a large umber of patiets, we used obesity (BMI > 25 kg/m 2 ) as a surrogate for high waist hip ratio, as oe of the compoet of the metabolic sydrome [20,21]. Statistical aalysis The results were expressed as media (rage) for cotiuous variables ad as frequecies (proportios) for categorical variables. Factors suspected to ifluece the severity of grade ad stage were tested i a uivariate ad multivariate logistic regressio aalysis. For the uivariate ad multivariate aalyses, the depedet variable was the presece of mild iflammatio (grades 0, 1) or severe iflammatio (grades 2, 3) for assessig the associatio with iflammatory grade ad mild fibrosis (stages 0, 1) or severe fibrosis (stages 2, 3, 4) for assessig associatio with fibrosis stage. The idepedet variables were age>35, sex, BMI > 25 kg/m 2, AST/ALT ratio >1, cholesterol >200 mg%, triglycerides >150 mg%, HOMA- IR >1.64, presece or absece of hypertesio ad metabolic sydrome. Correlatio betwee histological severity ad the above metioed idepedet variables was carried out usig Spearma s correlatio. For skewed data, o-parametric tests were used wherever ecessary. All data were aalyzed usig the SAS 8.0 statistical package. RESULTS From Jauary 1999 to Jue 2005, patiets with histologically prove NAFLD were registered at the liver cliic of

3 3402 ISSN CN / R World J Gastroeterol Jue 7, 2006 Volume 12 Number 21 AIIMS. This umber formed 2.6% of all liver cliic referrals durig this time period. The media age of this cohort of patiets was 34 (17-58) years, the BMI was 26.7 ( ) kg/m 2 ad 46 (90.1%) were males. Media AST/ALT ratio was 0.63 ( ). Table 1 gives the baselie demographic ad biochemical profile of the patiets. Prevalece of features of metabolic sydrome Table 2 depicts the prevalece of idividual features of the metabolic sydrome i these patiets. Eighty percet patiets had high HOMA-IR, while 69.4%, 40.8%, 36.4%, 11.8% ad 10% had obesity, hypertriglyceridemia, low HDL cholesterol, hypertesio ad diabetes respectively. All the ivestigatios ad cliical features required for diagosig the metabolic sydrome were available i 43 patiets. Of these, 9(20.9%) patiets fulfilled the criteria for a diagosis of metabolic sydrome (presece of 3 or more compoets of the metabolic sydrome). Histological severity ad factors associated with severe disease Twety-eight (55%) patiets had NASH (type 3 or 4 NAFLD) whereas the rest had oly fatty liver with or without spotty ecrosis without evidece of either ecrosis or fibrosis (type 1 or 2 NAFLD). Iflammatory grade: The majority of the patiets had mild iflammatio, either grade 1 [32 (63%)] or grade 2 [16 (31%)]. Oly 3(6%) patiets had severe or grade 3 iflammatio. We evaluated factors which could have bee associated with moderate to severe iflammatio (grade 2 or 3) versus those with mild iflammatio (grade 0 or 1). O uivariate aalysis, triglyceride levels more tha 150 mg % (OR = 7.1; 95% CI: , P = 0.002) ad AST/ALT ratio > 1 (OR = 14.3; 95% CI: , P = 0.008) were sigificatly associated with high grades of iflammatio. Presece of metabolic sydrome was more commo amog patiets who had severe iflammatio although it did ot reach the sigificat value (Table 3). O multivariate logistic regressio aalysis, hypertriglyceridemia more tha 150 mg% was the oly factor idepedetly associated with presece of high grade of iflammatio o idex biopsy (OR = 1.6; 95% CI: , P = 0.02). There was also positive correlatio betwee the triglyceride levels ad iflammatory grade (r = 0.412; P = 0.003) (Figure 1) Fibrosis stage: Most of the patiets had either o or miimal fibrosis. Twety-three (45%) ad 19 (37%) patiets had stage 0 ad stage 1 fibrosis respectively o liver biopsy. Eight (16%) had stage 2 ad 1 (2%) had stage 3 fibrosis. No patiet had cirrhosis o idex biopsy. O evaluatio of factors, which could have bee associated with severe fibrosis (stages 2, 3, 4 versus stages 0,1), metabolic sydrome (OR = 5.6; 95% CI: , P = 0.04) ad AST/ALT ratio > 1 (P = 0.09) were more commoly preset amog patiets with advaced fibrosis but failed to reach sigificace (Table 4). Table 1 Baselie demographic ad biochemical parameters of the cohort Parameters Age (yr) Media (Rage) Value 34 (16-58) Sex ratio (males/females) 46/5 Body mass idex (kg/m 2 ) Media (Rage) Bilirubi (mg%) Media (Rage) Aspartate amiotrasferase (IU/L) Media (Rage) Alaie amiotrasferase (IU/L) Media (Rage) Alkalie phosphatase (IU/L) Media (Rage) Albumi (gm%) Media (Rage) Cholesterol (mg%) Media (Rage) Triglycerides (mg%) Media (Rage) HDL cholesterol (mg%) Media (Rage) Fastig plasma glucose (mg%) Media (Rage) Fastig serum isuli (μiu/l) Media (Rage) HOMA-IR Media (Rage) 26.7 ( ) 0.8 ( ) 66 (29-230) 98 (52-3) 159 (87-372) (3-5.4) 180 (96-323) 145 (48-339) (28-65) 90 (70-150) ( ) ( ) DISCUSSION Although NAFLD has bee demostrated to be preset i almost oe quarter of the geeral populatio [10] ad has bee HDL: High desity lipoprotei, HOMA-IR: Homeostasis model assessmet for isuli resistace.

4 Mada K et al. Histological severity of NAFLD i Asia Idias 3403 Table 2 Prevalece of idividual compoets of metabolic sydrome Compoet of metabolic sydrome (%) Obesity 34 (69.4) Hypertriglyceridemia 20 (40.8) Low HDL cholesterol 16 (36.4) High HOMA-IR 36 (80) Hypertesio 6 (11.8) Diabetes 5 (9.8) Metabolic sydrome 9 (20.9) HDL: high desity lipoprotei; HOMA-IR: Homeostasis model assessmet for isuli resistace. Table 3 Factors associated with advaced grade of iflammatio Variables Grade 0/1 ( /N ) Grade 2/3 ( /N ) Male 29/32 17/ Age >35 years 13/32 10/ BMI >25 kg/m 2 18/30 16/ AST/ALT ratio > 1 1/32 6/ Cholesterol > 200 mg% 9/30 5/ Triglycerides >150 mg% 7/30 13/ HOMA-IR > /26 15/ Hypertesio 4/32 2/ Metabolic sydrome 3/30 6/ P 400 r = P = BMI: body mass idex; AST: Aspartate amiotrasferase; ALT: Alaie amiotrasferase; HOMA-IR: homeostasis model assessmet for isuli resistace; Uivariate aalyses; P value o chi-square or Fisher s exact test. Triglyceride levels (mg%) Iflammatory grade Figure 1 Scatter plot demostratig positive correlatio betwee triglyceride levels ad iflammatory grade. [Coefficiet of correlatio (r) = 0.412; P = usig spearma s correlatio]. show to be resposible for oe third of patiets with asymptomatic trasamiitis i Idia, little is kow about the baselie cliical, biochemical ad histological profile of these patiets. The preset study described the baselie profile i North Idia patiets with biopsy prove NAFLD. Our study demostrated NAFLD to be a disease of youger age group with a mea age at presetatio beig 33.3 (8.01) years. More tha 90% of the patiets were males. This predilectio for ivolvemet of males has bee see ot oly i other studies from Norther Idia [13-15], but also i the Uited States [22]. This is differet from what used to be the classical pheotype of NASH patiets, where it was described to be a disease of middle-aged females. The reaso for this male prepoderace has bee described as a higher waisthip ratio i me as compared to wome, a idicator of cetral obesity ad isuli resistace. Twety-oe percet of our patiets with NAFLD were foud to have the metabolic sydrome. The prevalece rate of metabolic sydrome i the geeral populatio of Idia has bee reported to be approximately 8% i me older tha 20 years [23], but o study has earlier reported the prevalece of the metabolic sydrome i patiets with NAFLD from Idia. The preset study also highlights certai importat differeces betwee the pheotype of the cohort of NAFLD as see i Norther Idia versus that which has bee de- Table 4 Factors associated with advaced stage of fibrosis Variables Stage 0/1 ( /N ) Stage 2/3/4 ( /N ) Male sex 4/42 1/9 1.0 Age >35 years 17/42 6/ BMI >25 kg/m 2 26/42 8/ AST/ALT ratio > 1 4/42 3/ Cholesterol > 200 mg% 13/40 1/ Triglycerides >150 mg% 14/40 6/ HOMA-IR > /36 8/ Hypertesio 5/42 1/9 1.0 Metabolic sydrome 5/40 4/ BMI: body mass idex; AST: Aspartate amiotrasferase; ALT: Alaie amiotrasferase; HOMA-IR: Homeostasis Model assessmet for isuli resistace; Uivariate aalyses; P value o chi-square ad fisher s exact test. scribed classically from the west. This differece would ot be surprisig, cosiderig the fact that the body fat distributio ad risk of cardiovascular complicatios with similar amout of body fat differs betwee Asia Idias ad their wester couterparts. This has eve prompted lowerig of BMI ad waist-hip ratio cut-offs for defiig overweight ad obesity amog Asias [24,25]. Sice obesity ad fat distributio are itegral parts of the isuli resistace sydrome ad fatty liver forms part of the spectrum of abormalities associated with this sydrome, the differece i the baselie profiles of NAFLD patiets would be obvious. The mea age of presetatio of our patiets, most of whom were detected durig workup of asymptomatic trasamiitis was about a decade youger tha the wester patiets. The youger age of presetatio may suggest icreasig health awareess amog our populatio with lower thresholds for health check-ups durig which the asymptomatic trasamiitis is picked up ad ivestigated. It may also suggest the P

5 3404 ISSN CN / R World J Gastroeterol Jue 7, 2006 Volume 12 Number 21 epidemic proportio of the metabolic sydrome i Idia, wherei the populatio which is fast acquirig wester lifestyle gets exposed to the detrimetal effects of high calorie diet ad physical iactivity at a earlier age. I fact, Idias have bee show to have higher body fat percetage ad adverse patter of body fat distributio icludig abdomial adiposity eve whe the BMI is withi limits cosidered as ormal for Caucasias [26]. Further, i the age group of years, the prevalece of overweight ad obesity is 11.63% ad 2.38% i urba ad 4.7% ad 3.63% i rural Idia populatio respectively [27]. The mea BMI of our patiets [26.5 (2.8) kg/m 2 ] was similar to cohorts described i other studies from North Idia [13,14], but sigificatly lower tha what has bee described i the Caucasia populatios, where the mea BMI has always bee reported to be above 30 [2,3,8,27,28]. As described earlier, Idia patiets despite havig a lower BMI have similar body fat percetage as the Caucasias. This would mea higher visceral fat distributio, evideced by higher waist hip ratios, despite lower BMIs. Ufortuately, waist hip ratios were ot available i the preset study for most of the patiets. However, 80% of our patiets had high HOMA-IR which is a idex of isuli resistace ad has bee correlated earlier with visceral adiposity ad higher waist hip ratios. Most importat fidig i the preset study was the presece of histologically mild disease i the cohort. Two thirds of the patiets had miimal or mild iflammatio (grade 0 or 1). Severe (grade 3) iflammatio was preset i oly 3 patiets. Hypertriglyceridemia was foud to be sigificatly ad idepedetly associated with presece of severe iflammatio. It has already bee demostrated that isuli resistace leads to higher free fatty acid load to the liver, cosequetly higher triglyceride sythesis ad icreased secretio of triglyceride rich very low desity lipoprotei (VLDL) from the liver [29,30]. I fact, circulatig VLDL triglycerides have bee correlated with liver fat desity [31]. Higher triglyceride ad fatty acid deposits may provide substrates for oxidative stress, which is a putative secod hit i pathogeesis of progressive NAFLD [32]. High triglyceride levels have bee show to be idepedetly associated with septal or advaced fibrosis by Ratziu et al. They also demostrated that septal fibrosis had a sigificat positive correlatio with liver ecroiflammatory activity, thereby idirectly likig hypertriglyceridemia with ecroiflammatio [9]. I the preset study however, we have demostrated a direct correlatio betwee hypertriglyceridemia ad ecroiflammatory activity i the liver. Iterestigly, oe of our patiets had cirrhosis o the idex biopsy ad oly 2% had advaced (stage 3 or 4) fibrosis. Most (> 90%) had miimal or o fibrosis o presetatio. Similar mild histology amog NASH patiets has bee reported i other studies from Idia [13-15] as well as oe study from Israel [33]. This is i cotrast to the reported series from the west, where advaced fibrosis or cirrhosis has bee demostrated to be preset i up to 50% of NAFLD cohorts [7-9,25]. Oe reaso which ca be cited for mild fibrosis i our patiets is the youger age at presetatio, because extet of fibrosis has bee show to icrease with icreasig age ad icreasig duratio of disease [8,9]. Therefore it is possible that the cohort i the preset study happeed to have bee detected at a early stage ad with advacig age the disease might progress. This ca oly be cofirmed later whe follow up histology is available i these patiets. Aother factor which could be resposible for milder disease i these patiets is the lower BMI. But this may ot be true, because as has bee discussed earlier, Asias with lower BMI may also be predisposed to similar risk of developig the metabolic sydrome as Caucasias with higher BMI. Further the effect of BMI or age was ot see o the degree of fibrosis i the uivariate aalysis. No other factor was foud to be associated with severe fibrosis i the preset study. The lack of associatio might be due to small umber of patiets i the group of severe fibrosis, as oly a miority of patiets had advaced fibrosis. Studies from the west, however have reported higher BMI, older age, type 2 diabetes mellitus, AST/ALT ratio > 1 ad hypertriglyceridemia to be idepedet risk factors for presece of advaced fibrosis o idex biopsy [8,9]. The milder histological disease i the preset cohort may also represet a geetic predispositio i our patiets, just like their geetic predispositio to develop metabolic sydrome. Whether these patiets with mild fibrosis cotiue to have a mild disease or they progress to a stage of cirrhosis ca oly be aswered oce we have a adequate follow up o them. To coclude, NAFLD i North Idia patiets is a disease of youg overweight males, most of whom are isuli resistat ad they ted to have a mild histological disease at presetatio. ACKNOWLEDGMENTS We sicerely ackowledge the valuable cotributios made by Dr Rajvir sigh (Seior scietist, Departmet of Biostatistics, All Idia Istitute of Medical Scieces, New Delhi) towards the statistical aalysis of this study. REFERENCES 1 Adams LA, Saderso S, Lidor KD, Agulo P. The histological course of oalcoholic fatty liver disease: a logitudial study of 103 patiets with sequetial liver biopsies. J Hepatol 2005; 42: Matteoi CA, Youossi ZM, Gramlich T, Boparai N, Liu YC, McCullough AJ. Noalcoholic fatty liver disease: a spectrum of cliical ad pathological severity. Gastroeterology 1999; 116: Lee RG. Noalcoholic steatohepatitis: a study of patiets. Hum Pathol 1989; 20: Bugiaesi E, Leoe N, Vai E, Marchesii G, Bruello F, Carucci P, Musso A, De Paolis P, Capussotti L, Salizzoi M, Rizzetto M. Expadig the atural history of oalcoholic steatohepatitis: from cryptogeic cirrhosis to hepatocellular carcioma.gastroeterology 2002; 123: Yoshioka Y, Hashimoto E, Yatsuji S, Kaeda H, Taiai M, Tokushige K, Shiratori K. Noalcoholic steatohepatitis: cirrhosis, hepatocellular carcioma, ad burt-out NASH. J Gastroeterol 2004; 39: Ludwig J, Viggiao TR, McGill DB, Oh BJ. Noalcoholic steatohepatitis: Mayo cliic experieces with a hitherto uamed disease. Mayo Cli Proc 1980; 55: Ratziu V, Giral P, Charlotte F, Bruckert E, Thibault V, Theodorou I, Khalil L, Turpi G, Opolo P, Poyard T. Liver fibrosis i overweight patiets. Gastroeterology 2000; 118: Sigh SP, Nayak S, Swai M, Rout N, Mallik RN, Agrawal O, Meher C, Rao M. Prevalece of oalcoholic fatty liver disease i coastal easter Idia: a prelimiary ultrasoographic survey. Trop Gastroeterol 2004; 25: Belletai S, Saccoccio G, Masutti F, Croce LS, Bradi G, Sasso F, Cristaii G, Tiribelli C. Prevalece of ad risk factors for

6 Mada K et al. Histological severity of NAFLD i Asia Idias 3405 hepatic steatosis i Norther Italy. A Iter Med 2000; 132: Clark JM, Bracati FL, Diehl AM. The prevalece ad etiology of elevated amiotrasferase levels i the Uited States. Am J Gastroeterol 2003; 98: Duseja A, Murlidhara R, Bhasali A, Sharma S, Das A, Das R, Chawla Y. Assessmet of isuli resistace ad effect of metformi i oalcoholic steatohepatitis- a prelimiary report. Idia J Gastroeterol 2004; 23: Paigrahi AK, Nada SK, Dixit RK, Acharya SK, Zuckerma AJ, Pada SK. Diagosis of hepatitis C virus-associated chroic liver disease i Idia: compariso of HCV atibody assay with a polymerase chai reactio for the 5' ocodig regio. J Med Virol 1994; 44: Stacy DL, Ha P. Serum ferriti measuremet ad the degree of agreemet usig four techiques. Am J Cli Pathol 1992; 98: Abe A, Yamashita S, Noma A. Sesitive, direct colorimetric assay for copper i serum. Cli Chem 1989; 35: Norma MR, Mowat AP, Hutchiso DC. Molecular basis, cliical cosequeces ad diagosis of alpha-1 atitrypsi deficiecy. A Cli Biochem 1997; 34: Brut EM, Jaey CG, Di Bisceglie AM, Neuschwader-Tetri BA, Baco BR. Noalcoholic steatohepatitis: a proposal for gradig ad stagig the histological lesios. Am J Gastroeterol 1999; 94: Chitturi S, Abeyguasekera S, Farrell GC, Holmes-Walker J, Hui JM, Fug C, Karim R, Li R, Samarasighe D, Liddle C, Weltma M, George J. NASH ad isuli resistace: Isuli hypersecretio ad specific associatio with the isuli resistace sydrome. Hepatology 2002; 35: Report of the expert committee o the diagosis ad classificatio of diabetes mellitus. Diabetes Care 1997; 20: Vikram NK, Misra A, Dwivedi M, Sharma R, Padey RM, Luthra K, Chatterjee A, Dhigra V, Jailkhai BL, Talwar KK, Guleria R. Correlatios of C-reactive protei levels with athropometric profile, percetage of body fat ad lipids i healthy adolescets ad youg adults i urba North Idia. Atherosclerosis 2003; 168: Executive summary of the third report of the Natioal Cholesterol Educatio Program (NCEP) Expert Pael o Detectio, Evaluatio, Ad Treatmet of High Blood Cholesterol i Adults (Adult Treatmet Pael III). Expert pael o detectio, evaluatio ad treatmet of high blood cholesterol i adults. J Am Med Assoc 2001; 285; Adams LA, Saderso S, Lidor KD, Agulo P. The histological course of o alcoholic fatty liver disease: a logitudial study of 103 patiets with sequetial liver biopsies. J Hepatol 2005; 42: Ruhl CE, Everhart JE. Determiats of the associatio of overweight with elevated serum alaie amiotrasferase activity i the Uited States. Gastroeterology 2003; 124: Camero AJ, Shaw JE, Zimmet PZ. The metabolic sydrome: prevalece i worldwide populatios. Edocriol Metab Cli North Am 2004; 33: World Health Orgaisatio, Wester Pacific regio. The Asia- Pacific perspective. Redefiig obesity ad its treatmet. WHO/IASO/IOTF, Appropriate body-mass idex for Asia populatios ad its implicatios for policy ad itervetio strategies. Lacet 2004; 363: Misra A, Vikram NK. Isuli resistace sydrome (metabolic sydrome) ad obesity i Asia Idias: evidece ad implicatios. Nutritio 2004; 20: Moha B, Kumar N, Aslam N, Ragbulla A, Kumbkari S, Sood NK, Wader GS. Prevalece of sustaied hypertesio ad obesity i urba ad rural school goig childre i Ludhiaa. Idia Heart J 2004; 56: Teli MR, James OF, Burt AD, Beett MK, Day CP. The atural history of oalcoholic fatty liver: A follow-up study. Hepatology 1995; 22: Lewis GF, Uffelma KD, Szeto LW, Weller B, Steier G. Iteractio betwee free fatty acids ad isuli i the acute cotrol of very low desity lipoprotei productio i humas. J Cli Ivest 1995; 95: Lewis GF, Steier G. Acute effects of isuli i the cotrol of VLDL productio i humas. Implicatios for the isuliresistat state. Diabetes Care 1996; 19: Baerji MA, Buckley MC, Chaike RL, Gordo D, Lebovitz HE, Kral JG. Liver fat, serum triglycerides ad visceral adipose tissue i isuli-sesitive ad isuli-resistat black me with NIDDM. It J Obes Relat Metab Disord 1995; 19: Marchesii G, Brizi M, Morselli-Labate AM, Biachi G, Bugiaesi E, McCullough AJ, Forlai G, Melchioda N. Associatio of oalcoholic fatty liver disease with isuli resistace. Am J Med 1999; 107: Kobler H, Schatter A, Zhoricki T, Malick SD, Keter D, Sokolovskaya N, Lurie Y, Bass DD. Fatty liver--a additioal ad treatable feature of the isuli resistace sydrome. Q J M 1999; 92: S- Editor Wag J L-Editor Zhu LH E- Editor Zhag Y