Nonalcoholic fatty liver disease in India a lot done, yet more required!

Transcription

1 Indian J Gastroenterol (2010) 29: DOI /s REVIEW ARTICLE Nonalcoholic fatty liver disease in India a lot done, yet more required! Ajay Duseja Received: 1 August 2010 / Accepted: 27 November 2010 / Published online: 30 December 2010 # Indian Society of Gastroenterology 2010 Abstract Nonalcoholic fatty liver disease (NAFLD) is emerging as an important cause of liver disease in India. Epidemiological studies suggest prevalence of NAFLD in around 9% to 32% of general population in India with higher prevalence in those with overweight or obesity and those with diabetes or prediabetes. Clinicopathological studies show that NAFLD is an important cause of unexplained rise in hepatic transaminases, cryptogenic cirrhosis and cryptogenic hepatocellular carcinoma in Indian patients. There is high prevalence of insulin resistance and nearly half of Indian patients with NAFLD have evidence of full-blown metabolic syndrome. Though oxidative stress is involved in the pathogenesis of NAFLD/ nonalcoholic steatohepatitis, serum or liver iron and HFE gene mutations appear not to play a role in the pathogenesis of NAFLD in Indian patients. Imaging modalities are not useful in differentiating simple steatosis from NASH and liver biopsy may be useful in those with risk factors for significant liver disease. Pilot studies on treatment strategies have shown that weight reduction and exercise, ursodeoxycholic acid, metformin, vitamin E and pentoxyfylline are effective in normalizing transaminases and or in improving hepatic steatosis and inflammation in Indian patients with NAFLD. Randomized controlled treatment trials involving large number of patients with histological end point are required to assess the efficacy of different modalities. In conclusion, a lot has been done, yet more is required to understand various aspects of NAFLD in India. A. Duseja (*) Department of Hepatology, Postgraduate Institute of Medical Education and Research, Chandigarh , India ajayduseja@yahoo.co.in Keywords Cirrhosis. Hepatocellular carcinoma. Insulin resistance. Metabolic syndrome. Nonalcoholic steatohepatitis. Type 2 diabetes mellitus Introduction Nonalcoholic fatty liver disease (NAFLD) is an entity that was described a few years ago. It includes patients with simple steatosis as also those with nonalcoholic steatohepatitis (NASH). NASH is a more advanced stage of NAFLD, and has a higher risk of progressing to liver cirrhosis or hepatocellular carcinoma (HCC) [1]. The term NASH was first introduced by Ludwig et al. in 1980 to describe histological changes indistinguishable from alcoholic hepatitis in patients with no or insignificant (less than 20 g/day) alcohol intake [2]. Even in the absence of alcohol intake, patients who have one or more components of the metabolic syndrome with insulin resistance, develop hepatic steatosis due to increased lipolysis and increased delivery of fatty acids from adipose tissue to liver [3]. Some of these patients with hepatic steatosis develop hepatic oxidative stress and recruitment of various cytokines, leading to hepatic inflammation and/or fibrosis and thus to NASH, setting the stage for future complications, such as cirrhosis and HCC. From the pathogenetic point of view, NAFLD has been classified as primary and secondary; primary NAFLD is usually associated with insulin resistance or metabolic syndrome, whereas secondary NAFLD is caused by intake of some drugs, surgery, or total parenteral nutrition [4]. Published literature on NAFLD from India is sparse. This may be related to (i) the fact that the condition was recognized fairly recently, (ii) a presumption that the condition is benign and has a non-progressive course, (iii) a large burden of viral hepatitis in India tends to reduce the priority accorded to this condition. In this article, I have

2 218 Indian J Gastroenterol (2010) 29: tried to review the published literature on primary NAFLD and NASH from India. Epidemiology of NAFLD Absence of signs and symptoms, and a lack of sensitive and specific diagnostic tests limit the ability to estimate the prevalence of NAFLD. The current epidemics of obesity and diabetes among adults and children residing in both developed and developing countries suggest that prevalence of NAFLD should have increased over time, and may be expected to increase further in future. NAFLD was initially described in obese, diabetic women, who denied alcohol use. Recent data indicate that NAFLD may have male preponderance or an equal gender distribution and may even occur in the absence of obesity or diabetes [4, 5]. NAFLD can occur at all ages including childhood, though the highest prevalence is described in those between years of age. With some limitations, both population and hospital-based studies from the West report that around 10 24% of general population, and 57 74% of obese individuals may have NAFLD. The corresponding rates for NASH are 3 4% and 15 20%, respectively [4, 5]. Urbanization and associated changes, such as sedentary life style and fat rich diet, and a higher inherited tendency for diabetes mellitus makes Indians more prone to metabolic syndrome or insulin resistance and its manifestations such as NAFLD and NASH [6 9]. Prevalence studies are available both in the general population and in persons with diabetes mellitus. NAFLD in general population Only a few studies have looked at the prevalence of NAFLD in the general population in India. A study from coastal regions of India found that 39 (24.5%) of 159 healthy attendants of patients had evidence of fatty liver on ultrasound (US) (males 27%, females 14%) [10]. Persons with ultrasonographic fatty liver had a higher body mass index (BMI) (mean 25.9 [4.2] kg/m 2 ) than those without (mean 22.1 [3.3] kg/m 2 ). The study concluded that NAFLD may be as common in the developing world as in developed countries despite a lower prevalence of obesity [10]. In another study, residents of two railway colonies at Mumbai were evaluated clinically and by abdominal US for the presence of NAFLD [11]. Among the 1,168 participants, overall prevalence of NAFLD on US was 16.6%, being higher in males than in females (24.6% vs. 13.6%). Risk factors associated with NAFLD were age above 40 years, male gender, central obesity, high BMI (>25), elevated fasting blood sugar, and high AST and ALT levels [11]. In a more detailed study, prevalence of NAFLD and its association with glucose intolerance (type 2 diabetes mellitus [DM], prediabetes or metabolic syndrome) as defined by the modified Adult Treatment Panel III (ATP III) criteria was studied in 541 urban southern Indian subjects [12]. Anthropometry and lipid estimations were done in all subjects, and oral glucose tolerance test in all, except self-reported diabetic subjects. NAFLD was diagnosed by US and metabolic syndrome by ATP III criteria. The overall prevalence of NAFLD was 32% (173/541 subjects) (men: 35.1%, women: 29.1%). NAFLD was more frequent (54.5%) in subjects with DM than in those with prediabetes (impaired glucose tolerance [IGT] or impaired fasting glucose [IFG]) (33%), isolated IGT (32.4%), isolated IFG (27.3%) and normal glucose tolerance (NGT) (22.5%). Even after adjustment for age, gender and waist circumference, NAFLD was associated with diabetes and metabolic syndrome [12]. In a more recent, community-based epidemiological study, 1,911 inhabitants of a rural area in West Bengal were studied [13]. Rather than NAFLD, the authors used the term nonalcoholic fatty liver (NAFL) and the diagnosis was based on a dual radiological screening protocol consisting of ultrasonographic and computed tomographic examination of the liver. Transient elastographic examination and liver biopsy were performed in a subset to identify significant liver disease. The prevalence rates of NAFL, NAFL with elevated alanine aminotransferase, and cryptogenic cirrhosis were 8.7%, 2.3%, and 0.2%, respectively. The risk of NAFL was the highest in those with BMI >25 kg/m 2. Abdominal obesity, dysglycemia (fasting plasma glucose >100 mg/dl or elevated homeostatic model assessment of insulin resistance [HOMA-IR]), and higher income were other risk factors. Surprisingly, normal BMI ( kg/m 2 ) was associated with a 2-fold increased risk of NAFL than in those with a BMI <18.5 kg/m 2 [13]. Diabetes mellitus Type 2 DM is a major risk factor for presence and severity of NAFLD. Conversely, only a minority of Indian subjects with NAFLD and raised transaminases have diabetes at presentation. We found that 35 of our 40 (88%) nonalcoholic patients with DM had evidence of fatty liver on US; liver biopsy was not done [14]. In a study from Mumbai, 49 of 100 patients with DM had evidence of fatty liver on US; of these 32 underwent liver biopsy [15]. Of these biopsies, 21 (66%), 4 (13%) and 3 (9%) showed mild, moderate and severe NASH, respectively, and 7 (22%) showed fibrosis. There was no difference in body mass index (BMI), transaminase levels, and serum cholesterol and triglyceride levels between diabetic patients with and without NAFLD [15].

3 Indian J Gastroenterol (2010) 29: In another study, 47 randomly-selected patients of DM were screened, and those without other causes of liver disease (viral, alcoholic, drug, autoimmune, etc.) were investigated for NAFLD using hepatic US and liver biopsy [16]. Histology was normal in 17%, only fatty change was seen in 43%, NASH in 40% with more advanced disease in 23% amongst NASH patients with low prevalence of cirrhosis (2.12%). Risk factors for NASH included positive family history of DM and hypertension, presence of hypertension, longer duration of diabetes, female sex and high body mass index [16]. NAFLD as a cause of significant liver disease Available evidence indicates that NAFLD is responsible for a large proportion of persons presenting with unexplained transaminase elevation, and that it is an important cause of cryptogenic cirrhosis and cryptogenic HCC. NAFLD and unexplained elevation of transaminases In a study from New Delhi, NASH and chronic viral hepatitis were the most common causes of asymptomatic rise in hepatic transaminases [17]. NAFLD is the third most common cause of chronic liver disease (constituting around 20%) in our experience, after alcoholic liver disease (~40%) and chronic viral hepatitis (~38%) [18]. Though NASH is a more commonly used term, it is preferable to restrict this term to only those with histological features of NASH rather than to all persons with NAFLD and raised transaminase levels. Overall, serum transaminase levels have a poor correlation with histological severity in patients with NAFLD. Patients with raised ALT levels may not have histological NASH, and normal ALT levels do not preclude severe disease. Amarapurkar et al. found histological evidence of cirrhosis in 23% of their 25 patients with NASH and normal ALT [19]. In fact, in this study, the clinical and histological spectrums of NAFLD were no different between 25 patients with normal ALT and 56 patients with abnormal ALT. NAFLD and cryptogenic cirrhosis NAFLD/NASH has been shown to be the most common cause of cryptogenic cirrhosis in many studies from the West. As with other chronic liver diseases, NAFLD/NASH may be asymptomatic initially and may present later as cryptogenic cirrhosis or HCC. Since NAFLD/NASH was recognized as a clinical entity only recently, patients who have had unrecognized NAFLD/NASH in the past may present as cirrhosis of unknown etiology in the future. It may be difficult to recognize NAFLD/NASH in such patients even at histology, since liver fat decreases as fibrosis increases. Thus, characteristic changes of NAFLD/ NASH may not be evident once cirrhosis supervenes. In a study from Mumbai, NASH with or without cirrhosis, and cryptogenic cirrhosis occurred more commonly in patients with diabetes than those without it [20]. The frequency of diabetes was 57% in patients with cryptogenic cirrhosis and 30% in those with noncryptogenic cirrhosis, suggesting an association between diabetes mellitus and liver cirrhosis [20]. In our 25 patients with cryptogenic cirrhosis and 18 patients with cirrhosis related to viral hepatitis and other causes, patients with cryptogenic cirrhosis had a higher prevalence of DM (44% vs. 11%), higher BMI (mean 26 kg/m 2 vs. 23 kg/m 2 ) and higher frequency of abnormal levels of high density lipoproteins (HDL) (40% vs. 6%). These findings suggest that NAFLD/NASH may be a cause of cryptogenic cirrhosis [21]. In a recent study from a liver transplant center, clinicopathological features of NAFLD were explored by the clinical data and by examining explant livers in 103 living-donor liver transplant recipients [22]; among these, 30 had a pre-liver transplant diagnosis of cryptogenic cirrhosis. Nineteen of them (63%) were finally labeled as NAFLD-related cirrhosis and showed histological features that differed in several respects from those reported for early and established phases of NAFLD. It was concluded that when appropriate criteria are used, NAFLD appears to account for close to two thirds of cases currently labeled as cryptogenic cirrhosis [22]. We recently studied surrogate markers of NAFLD in 65 patients with cryptogenic cirrhosis and compared these with 50 patients with virus-related cirrhosis of comparable age, gender and severity of liver disease [23]. Among patients with cryptogenic cirrhosis, viral causes, autoimmune disease, Wilson s disease and iron overload were excluded; in addition, occult HBV infection (using total anti-hbc) and celiac disease (using anti-tissue transglutaminase antibodies, anti-endomysial antibodies and duodenal biopsy) were excluded in 16 and 10 patients, respectively. Patients with cryptogenic cirrhosis had higher mean BMI (26.1 [6.0] kg/m 2 vs [1.7] kg/m 2 ), and were more likely to have abnormal waist circumference (38/65 [59%] vs. 15/50 [30%]), type 2 DM (26 [40%] vs. 5 [10%]) and low serum HDL (35 [54%] vs. 3 [6%]) [23]. NAFLD in cryptogenic hepatocellular carcinoma (HCC) We analysed surrogate markers of NAFLD in 39 patients with cryptogenic HCC (known etiologies for HCC excluded) and compared the results with 39 patients

4 220 Indian J Gastroenterol (2010) 29: with virus-related HCC [23]. Patients with cryptogenic HCC had a higher BMI (24.4 [4.0] kg/m 2 vs [3.4] kg/m 2 ) and a higher prevalence of DM (15 [39%] vs. 7 [18%]) than those with virus-related HCC. There was no difference in proportion of patients with abnormal HDL levels, abnormal serum triglyceride levels and hypertension in the two groups. Using metabolic risk factors as surrogate markers of NAFLD, it appears that NAFLD is an important cause of both cryptogenic cirrhosis and cryptogenic hepatocellular carcinoma HCC in India [23]. Pathogenesis Pathogenesis of the NASH is not completely understood. The first step appears to involve deposition of excess fat in the liver; this is followed in some patients by increased fatty acid oxidation, oxidative stress and cytokine production, resulting in progression to steatohepatitis and fibrosis. Various pathogenetic mechanisms that pay a role include cytokines (tumor necrosis factor (TNF)-α, adiponectin, resistin, leptin, interleukins, transforming growth factor ß etc.) that lead to insulin resistance, and serum and liver iron overload and oxidative stress that lead to necroinflammation and fibrosis [3 5]. Insulin resistance Insulin resistance is believed to be the key factor that leads to increased lipolysis in peripheral adipose tissue and increased uptake of fatty acids by hepatocytes. Hyperinsulinemia resulting from insulin resistance also adds to fatty acid content of hepatocytes by increasing glycolysis and by decreasing apolipoprotein B-100 production, and hence export of fatty acids as very low density lipoproteins (VLDL). The end result is an increase in fatty acids and triglycerides in the hepatocytes leading to steatosis. Insulin resistance is almost universal in patients with NAFLD and is related to an imbalance between proinsulin (adiponectin) and anti-insulin cytokines (TNF-α), particularly those secreted from adipose tissue (adipokines) [24]. Among our patients, 83% to 98% had evidence of insulin resistance using insulin tolerance test (ITT) and HOMA IR [25 27]. Studies from other Indian centers have also reported insulin resistance to be common in patients with NAFLD [28 30]. Though glucose clamp studies are ideal method of studying insulin resistance, most studies in Indian patients with NAFLD have used HOMA-IR. We compared ITT with HOMA-IR in a subset of 22 patients with NAFLD and found a good correlation between their results [31]. Metabolic syndrome Currently, NAFLD is considered to be an integral part of the metabolic syndrome with insulin resistance as a central pathogenic factor. Metabolic syndrome is characterized by the presence of insulin resistance in association with other metabolic abnormalities such as obesity, diabetes, dyslipidemia and hypertension. According to adult treatment panel III (ATP III) criteria, metabolic syndrome is defined by presence of at least 3 of the 5 criteria, namely obesity, diabetes mellitus, hypertension, low HDL, high triglycerides. We found full-blown metabolic syndrome as defined by the ATP III criteria in 50% of our patients [26, 27], and at least one of the above criteria in 90% of patients (unpublished data). Other centers from India have also reported the presence of metabolic syndrome in 21 68% of patients, and at least one criterion in almost all the patients [28, 29]. Lower prevalence of full-blown metabolic syndrome in Indian patients with NAFLD may be related to the lower frequency of DM and hypertension in patients presenting with raised transaminases [26]. In a recent casecontrol study, Asian Indians with NAFLD were shown to have higher fasting plasma insulin levels, and higher frequency of metabolic syndrome (41% among cases and 20% in controls) and impaired fasting glucose (23% among cases and 5% among controls), than those without NAFLD. A step-wise logistic regression analysis showed BMI, fasting blood glucose and fasting insulin level to be independent predictors of NAFLD [32]. Overweight/obesity Initial studies from India that used the international criteria [33] for defining overweight and obese states reported obesity in only 12 30% of patients with NAFLD [34, 35]. We found overweight and obesity in 64% and 12% of our patients using the international criteria [25]. However, our subsequent studies showed these frequencies to be higher (20% overweight and 68% obese) when we used the Asia-Pacific criteria [36] for these states [26, 27]. Asians have a high percentage of body fat compared to white Caucasians and blacks despite having lower average BMI [37]. For a particular percentage of body fat, BMI values of Asians including Asian Indians are 3 kg/m 2 lower than those in Caucasians [38]. This difference is partly explained by the body build (trunk to leg-length ratio), low muscularity, adaptation to chronic calorie deprivation, and ethnicity [39]. More importantly, the morbidity and mortality associated with higher body fat occur more frequently at lower BMI in Asians than in Caucasians. In a study from Korea, optimum BMI cut-off to predict diabetes, hypertension and dyslipidemia was found to be lower than the internationally recommended cut-off [40]. A study from Delhi showed that about 66% of men and 88% of women classified as non-obese based on international cut-off of

5 Indian J Gastroenterol (2010) 29: BMI had one or more cardiovascular risk factors [39]. Based on these data, it has been suggested that the BMI limits for overweight and obesity should be lower for Asian Indians [36, 41]. In a recent consensus meeting in Delhi, optimum cut-offs for obesity (BMI and waist circumference) and metabolic syndrome were adopted [42]. Recommendations for BMI are: normal kg/m 2, overweight kg/m 2 and obesity as BMI 25 kg/m 2 [42]. Central or abdominal obesity that is more commonly associated with insulin resistance and has been observed in 80 90% of Indian patients with NAFLD [26 30]. A high prevalence of abdominal obesity is seen in Asians, including Asian Indians even when the BMIislessthan25kg/m 2 [43]. Similarly Asians have been found to have more intra-abdominal adipose tissue than white Caucasians, in spite of having smaller waists [43]. Lower cutoffs (waist circumference 90 cm in males and 80 cm in females) are also recommended for identifying abdominal obesity in Asian Indians [42]. Diabetes mellitus NAFLD has been associated very closely with the presence of type 2 diabetes mellitus. DM is an important determinant of both presence and severity of NAFLD [15, 16]. We found that DM and impaired glucose tolerance was uncommon in patients of NAFLD presenting with raised transaminases, being present in only 12% and 14% of our patients, respectively; however, IR was found in more than 80% of patients [26, 27]. We hypothesized that milder degree of IR in our patients with NAFLD may be adequate to cause fatty liver, but may not cause frank DM [26, 27]. Other centers have also reported a lower prevalence of DM at presentation in Indian patients with NAFLD [28 30]. Hypertension As with DM, hypertension is also not common in Indian patients with non-cirrhotic NAFLD presenting with raised transaminases. It was observed in 10% to 12% of patients in our study and from New Delhi [26 28]. Dyslipidemia In contrast to low prevalence of DM and hypertension, dyslipidemia is a common feature, and is present in approximately 50% of Indian patients with NAFLD [26, 27]. Both components of metabolic syndrome (high triglycerides and low HDL) were observed with almost equal frequency, being present in 53% and 66% patients with NAFLD, respectively [26]. Oxidative stress An increased load of fatty acids in hepatocytes increases the rate of mitochondrial β-oxidation and levels of cytochrome P450 4A and cytochrome P4502E1, leading to increase in reactive oxygen species. The increased mitochondrial oxidative stress provides the second hit, facilitating progression from steatosis to steatohepatitis and fibrosis through three main mechanisms, namely lipid peroxidation, cytokine induction, and Fas ligand induction. A high ratio of reduced glutathione (GSH) to oxidized glutathione (GSSG) protects against oxidative stress. In patients with NAFLD, hepatic GSH levels are reduced, but the redox imbalance may be lower than that seen in patients with alcoholic liver disease [44 47]. In a study from New Delhi, oxidative stress and antioxidant status were studied in 29 prospectively enrolled patients with NAFLD, 25 disease controls with chronic viral hepatitis, and 23 healthy controls [44]. Thiobarbituric acid reactive substances (TBARS), which represent products of lipid peroxidation, were significantly increased among patients with NAFLD than in the other two groups [44]. In another study by the same group, patients with alcoholic liver disease were found to have greater degree of redox imbalance than patients with NAFLD [45]. To explore the relationship of oxidative stress and insulin resistance in NAFLD subjects with and without type 2 DM, 200 subjects were recruited from the Chennai Urban Rural Epidemiology Study group, and levels of TBARS, protein carbonyl (PCC) and glutathione levels were measured [46]. Levels of TBARS and PCC were significantly higher and GSH/GSSG ratio was significantly lower in diabetic subjects with NAFLD than in other groups, suggesting that oxidative stress markers are significantly associated with NAFLD after adjustment for age, gender, BMI and glycemic status [46]. In another study from southern India, measures of oxidative stress correlated with clinical findings in 35 NAFLD patients, 38 alcoholic liver disease patients and 38 normal subjects. Glutathione content, catalase activity, glutathione reductase activity and glutathione peroxidase activity were found to be reduced in patients with NAFLD [47]. Iron/HFE gene mutations The relationship between iron overload and NAFLD is complex and remains unresolved, with studies both favoring and refuting a role for iron in the pathogenesis of NAFLD. Saturation of mitochondrial β-oxidation leads to peroxisomal oxidation and generation of hydrogen peroxide, which in the presence of increased iron is converted to hydroxyl radicles, adding to the oxidative stress and cellular injury. Among our 60 patients, abnormal serum iron, ferritin and transferrin saturation were found in only 3%, 7% and 8%, respectively; no patient had C282Y HFE gene mutation and only 4 (13%) had heterozygous H63D gene mutation among 30 patients [48, 49]. These findings

6 222 Indian J Gastroenterol (2010) 29: suggest that iron abnormalities do not play much role in the pathogenesis of NAFLD in Indian population. It also indicates that presence of C282Y HFE gene mutation may be related to the ethnicity of the population, and that of H63D mutation may be nonspecific. A majority of our patients (67%) had negative Perl s staining for iron on liver biopsy, with only 20% and 13% showing 1+ or 2+ staining, supporting a lack of any role for iron in our patients with NAFLD [48, 49]. Some of the recent work from India on the pathogenesis of NAFLD has suggested derangements in hepatic gluconeogenesis, using non-invasive assessment by 31 P magnetic resonance (MR) spectroscopy [50]. Another recent study has identified a nonhuman primate model of NAFLD resembling features of human NAFLD [51]. This will be useful in understanding the mechanism of the onset of this disease and for developing novel therapeutic modalities. Diagnosis Clinical features and imaging in NAFLD Most patients with non-cirrhotic NAFLD are initially asymptomatic, and are diagnosed due to an incidental detection of raised liver enzymes or fatty liver on US. In some patients, these findings are detected during work up for dyspeptic symptoms, malaise or fatigability, or during work up for other illnesses. Of our 100 patients, 37 presented with dyspepsia, 36 had malaise or fatigability, and 22 had incidental detection of raised transaminases and fatty liver on US during work up for organ donation, an illness or executive health check [26]. Five patients who presented with intermittent jaundice were later found to have associated Gilbert s syndrome on investigations [52]. Anthropometry may reveal overweight or obesity. Mild hepatomegaly may be an important sign in around half of these patients but signs of liver failure are absent unless patient has progressed to cirrhosis or HCC. Biochemical tests usually show mildly elevated AST and ALT, with the latter often higher. US has a sensitivity and specificity of around 85% in detecting steatosis; CT and MR imaging have no advantage over US for detecting liver fat. None of these tests can differentiate between steatosis alone and histological NASH. We studied 10 patients, including 5 with histological NASH, using both US and chemical shift MR imaging [53]. Though MR was more objective than US in picking up steatosis, neither test could pick up hepatic fibrosis and differentiate patients with NASH from those without NASH [53]. MR spectroscopy may be somewhat better in detecting fat, but cannot detect the degree of inflammation and fibrosis. Fibroscan (tissue elastography) is a new non-invasive modality in detecting liver fibrosis. Its role in various liver diseases including NAFLD is still evolving. Till then, liver biopsy remains the only method for detection of fat and associated necro-inflammation and fibrosis (NASH) in the liver, and for grading and staging the disease in NAFLD. NAFLD and liver biopsy Even though a definite diagnosis of NAFLD/NASH can be made only on histology, convincing patients to undergo a liver biopsy is difficult due to the slowly progressive nature of the disease and lack of specific treatment. We could do liver biopsy in 43 of 127 patients with NAFLD who presented with persistent raised ALT [27]. In spite of a selection bias, clinical characteristics of 43 biopsy proven patients were similar to 84 non-biopsy proven patients. A recent study from Delhi described typical histological features of NAFLD as the presence of macrovesicular steatosis, lobular neutrophilic inflammation, presence of Mallory bodies, ballooning degeneration, lipogranuloma and pericellular fibrosis. These changes predominate in perivenular regions, i.e., zone 3 of hepatic acinus [54]. In a separate study, female gender, BMI, waist-to-hip ratio, hypercholesterolemia and LDL levels were found to be independent predictors of disease severity in patients with NASH [55]. Since this subgroup of patients with NAFLD would be expected to derive the most benefit from therapy, they are possibly the best candidates for liver biopsy. Overall, liver biopsies in Indian patients presenting with raised transaminases show only mild to moderate degree of inflammation and mild to moderate stage of fibrosis. Histological NASH is found in only about half of them, and cirrhosis at presentation is uncommon [26 28]. Liver biopsy confirmed the diagnosis of NAFLD in all; histological NASH was seen in only 22 of our 43 patients. Amongst 22 patients with NASH, all had mild (10 patients) or moderate (12 patients) inflammation. Six patients did not have fibrosis, whereas stage 1, 2 and 3 fibrosis was seen in 7, 5 and 4 patients, respectively with none of the patients having stage 4 (cirrhosis) fibrosis [27]. Similar results were reported in another study from New Delhi [28]. Treatment The pharmacological treatment of patients with NAFLD is still evolving. No single therapy for NAFLD has been clearly proven to be effective. Various treatment modalities used for NAFLD have included lifestyle modifications such as weight loss and exercise, treatment of risk factors like diabetes mellitus and hyperlipidemia, and use of insulin sensitizing agents such as biguanides (metformin), thiazo-

7 Indian J Gastroenterol (2010) 29: lidinediones (rosiglitazone, pioglitazone) and antioxidants and various hepatoprotective agents. A recent meta-analysis of randomized trials in the treatment of NAFLD found that weight loss, thiazolidinediones (especially pioglitazone), and antioxidants have been most extensively evaluated [56]. Weight loss was found to be safe and dose-dependently improved histological disease activity in NASH, but more than 50% of patients failed to achieve target weight loss. Thiazolidinediones improved steatosis and inflammation but yielded significant weight gain. Randomized controlled trials with antioxidants yielded conflicting results and were heterogeneous with respect to type and dose of drug, duration and implementation of lifestyle intervention. Among other agents, pentoxifylline, telmisartan and L-carnitine improved liver histology in at least 1 randomized trial in NASH [56]. In a recent study from Lucknow, at least 30 min of aerobic exercise daily was found to be effective in reducing weight and normalizing serum transaminases [30]. Around three-fourths (74%) of our patients showed biochemical response to lifestyle modifications and ursodeoxycholic acid (UDCA) [26]; however this study did not have a control group and thus the improvement may not be related to these interventions. There are no studies from India on the use of thiazolidinedione group of drugs. In an open label study of metformin in 17 patients with NAFLD [25, 26], none had any side effects and more than half (59%) showed normalization of ALT [26]. However, this study too lacked a control group, making it difficult to ascribe the benefit to metformin. In another study, metformin was found to be effective in 25 patients with NAFLD who did not respond to lifestyle interventions [57]. In a retrospective study from Delhi, a combination of life style modifications, UDCA and vitamin E was found to be superior to either lifestyle modifications alone or a combination of lifestyle modifications and UDCA in normalizing ALT in patients with NASH [58]. All 12 patients in the vitamin E group normalized their ALT levels in comparison to 8 of 18 patients and 5 of 12 patients in the lifestyle modifications and UDCA group respectively (p=0.003). In an uncontrolled study, Satapathy et al. used pentoxyfylline (400 mg tid for 6 months), which has anti-tnf activity, in 20 patients with NASH and reported biochemical response and improvement in insulin resistance [59]. In a recent study, the same group also showed the histological improvement in steatosis, lobular inflammation and fibrosis in 9 patients with NASH who received pentoxyfylline for 12 months [60]. Large, multicenter, randomized placebo-controlled trials using various treatment modalities are needed to determine the optimum treatment. Experimental treatments, such as personalized depot-specific liposuction to remove the inactive adipocytes followed by adipocyte repopulation in the adipose tissue, are currently being studied [61]. Are there any racial differences? Because of ethnic differences in the occurrence of metabolic syndrome and its components, effect of ethnicity on clinicopathological profile of NAFLD patients has been studied [39]. One such study showed that the prevalence of insulin resistance, defined as the lower quartile of insulin sensitivity index, was approximately 2- to 3-fold higher in Asian-Indians than in other ethnic groups [62]. Further, Asian-Indian men had an approximately 2-fold higher hepatic triglyceride content and plasma IL-6 concentrations than Caucasian men [62]. Majority of Indian patients with NAFLD are overweight or obese as per the Asian-Pacific criteria even though they do not have morbid obesity that is seen in Western patients [63]. Other differences in Indian patients include a lower frequency of metabolic syndrome including its components such as diabetes mellitus and hypertension, lower frequencies of iron abnormalities and HFE gene mutations, and presence of a histologically milder disease at presentation [63]. However, data on the natural history of NAFLD and the rates of serious significant chronic liver disease in such patients are lacking. Thus, there is a clear need for further prospective, cooperative studies [64]. In conclusion, NAFLD is responsible for significant liver disease in India. A lot has been done regarding the epidemiology, pathogenesis, diagnosis and treatment of patients with NAFLD. Yet more is required, to complete the various missing links. References 1. Falck-Ytter Y, Younossi ZM, Marchesini G, Mc Cullough AJ. Clinical features and natural history of nonalcoholic steatosis syndromes. Semin Liver Dis. 2001;21: Ludwig J, Viggiano T, McGill D, Ott B. Nonalcoholic steatohepatitis. Mayo Clinic experience with a hitherto unnamed disease. Mayo Clin Proc. 1980;55: Angulo P. Nonalcoholic fatty liver disease. N Engl J Med. 2002;346: AGA technical review on nonalcoholic fatty liver disease. Gastroenterology. 2002;123: Neuschwander-Tetri BA, Caldwell SH. Nonalcoholic steatohepatitis: summary of an AASLD single topic conference. Hepatology. 2003;37: Misra A, Vikram NK. Insulin resistance syndrome (metabolic syndrome) and obesity in Asian Indians: evidence and implications. Nutrition. 2004;20: Misra A, Luthra K, Vikram NK. Dyslipidemia in Asian Indians: determinants and significance. J Assoc Physicians India. 2004;52: Ramachandran A, Snehalatha C, Satyavani K, Vijay V. Impaired fasting glucose and impaired glucose tolerance test in urban population in India. Diabet Med. 2003;20: Duseja A, Chawla Y. Nonalcoholic fatty liver disease in India: how much? How soon? Trop Gastroenterol. 2005;26:1 3.

8 224 Indian J Gastroenterol (2010) 29: Singh SP, Nayak S, Swain M, et al. Prevalence of nonalcoholic fatty liver disease in coastal eastern India: a preliminary ultrasonographic survey. Indian J Gastroenterol. 2004;25: Amarapurkar D, Kamani P, Patel N, et al. Prevalence of nonalcoholic fatty liver disease: population based study. Ann Hepatol. 2007;6: Mohan V, Farooq S, Deepa M, Ravikumar R, Pitchumoni CS. Prevalence of non-alcoholic fatty liver disease in urban south Indians in relation to different grades of glucose intolerance and metabolic syndrome. Diabetes Res Clin Pract. 2009;84: Das K, Das K, Mukherjee PS, et al. Nonobese population in a developing country has a high prevalence of nonalcoholic fatty liver and significant liver disease. Hepatology. 2010;51: Duseja A, Bhansali A, Bhadada S, Jain P, Dhiman RK, Chawla Y. Nonalcoholic fatty liver disease in patients with recent onset type 2 diabetes mellitus (abstract). J Gastroenterol Hepatol. 2004;19 Suppl:A Gupte P, Amarapurkar D, Agal S, et al. Non alcoholic steatohepatitis in type 2 diabetes mellitus. J Gastroenterol Hepatol. 2004;19: Banerjee S, Ghosh US, Dutta S. Clinicopathological profile of hepatic involvement in type-2 diabetes mellitus and its significance. J Assoc Physicians India. 2008;56: Madan K, Batra Y, Panda SK, et al. Role of polymerase chain reaction and liver biopsy in the evaluation of patients with asymptomatic transaminitis: Implications in diagnostic approach. J Gastroenterol Hepatol. 2004;19: Dhiman RK, Duseja A. Nonalcoholic Fatty Liver Disease. In. J Assoc Physicians India. Medicine Update. Eds. Gupta SB. 2005;15: Amarapurkar DN, Patel ND. Clinical spectrum and natural history of non-alcoholic steatohepatitis with normal alanine aminotransferases values. Trop Gastroenterol. 2004;25: Amrapurkar D, Das HS. Chronic liver disease in diabetes mellitus. Trop Gastroenterol. 2002;23: Duseja A, Nanda M, Das A, Das R, Bhansali A, Chawla Y. Prevalence of obesity, diabetes mellitus and hyperlipidemia in patients with cryptogenic liver cirrhosis. Trop Gastroenterol. 2004;25: Nayak NC, Vasdev N, Saigal S, Soin AS. End-stage nonalcoholic fatty liver disease: evaluation of pathomorphologic features and relationship to cryptogenic cirrhosis from study of explant livers in a living donor liver transplant program. Hum Pathol. 2010;41: Duseja A, Sharma BK, Kumar A, et al. Nonalcoholic fatty liver in a developing country is responsible for significant liver disease. Hepatology. 2010;52: Tilg H, Diehl AM. Cytokines in alcoholic and nonalcoholic steatohepatitis. N Engl J Med. 2000;343: Duseja A, Murlidharan R, Bhansali A, et al. Assessment of insulin resistance and effect of metformin in nonalcoholic steatohepatitis a preliminary report. Indian J Gastroenterol. 2004;23: Duseja A, Das A, Das R, et al. Clinicopathological profile of Indian patients with nonalcoholic fatty liver disease is different from that in the west. Dig Dis Sci. 2007;52: Duseja A, Das A, Dhiman RK, et al. Indian patients with nonalcoholic fatty liver disease presenting with raised transaminases are different at presentation. World J Gastroenterol. 2007;13: Madan K, Batra Y, Gupta SD, et al. Non-alcoholic fatty liver disease may not be a severe disease at presentation among Asian Indians. World J Gastroenterol. 2006;12: Amarapurkar DN, Patel ND. Prevalence of metabolic syndrome in non-diabetic and non-cirrhotic patients with non-alcoholic steatohepatitis. Trop Gastroenterol. 2004;25: Sreenivasa Baba C, Alexander G, Kalyani B, et al. Effect of exercise and dietary modification on serum aminotransferase levels in patients with nonalcoholic steatohepatitis. J Gastroenterol Hepatol. 2006;21(1 Pt 1): Duseja A, Thumburu KK, Das A, et al. Insulin tolerance test is comparable to homeostasis model assessment for insulin resistance in patients with nonalcoholic fatty liver disease. Indian J Gastroenterol. 2007;26: Bajaj S, Nigam P, Luthra A, et al. A case-control study on insulin resistance, metabolic co-variates and prediction score in nonalcoholic fatty liver disease. Indian J Med Res. 2009;129: Willett WC, Dietz WH, Colditz GA. Guidelines for healthy weight. N Engl J Med. 1999;341: Amarapurkar DN, Amarapurkar AD. Nonalcoholic steatohepatitis: clinicopathological profile. J Assoc Physicians India. 2000;48: Agarwal SR, Malhotra V, Sakhuja P, Sarin SK. Clinical, biochemical and histological profile of nonalcoholic steatohepatitis. Indian J Gastroenterol. 2001;20: Steering Committee of the Western Pacific region of the World Health Organization. The International Association for the study of Obesity and the International Study Task Force. The Asia- Pacific perspective; redefining obesity and its treatment. Health Communications, Australia. 2000; Dudeja V, Misra A, Pandey RM, Devina G, Kumar G, Vikram NK. BMI does not accurately predict overweight in Asian Indians in northern India. Br J Nutr. 2001;86: Deurenberg-Yap M, Chew SK, Deurenberg P. Elevated body fat percentage and cardiovascular risks at low body mass index levels among Singaporean Chinese, Malays and Indians. Obes Rev. 2002;3: Misra A, Misra R. Asian Indians and insulin resistance syndrome: global perspective. Metab Syndr Relat Disord. 2003;1: Ko GT, Chan JC, Cockram CS, Woo J. Prediction of hypertension, diabetes, dyslipidaemia or albuminuria using simple anthropometric indexes in Hong Kong Chinese. Int J Obes Relat Metab Disord. 1999;23: Dhiman RK, Duseja A, Chawla Y. Asians need different criteria for defining obesity. Arch Intern Med. 2005;165: Misra A, Chowbey P, Makkar BM, et al. Consensus statement for diagnosis of obesity, abdominal obesity and the metabolic syndrome for Asian Indians and recommendations for physical activity, medical and surgical management. J Assoc Physicians India. 2009;57: Misra A, Vikram NK. Clinical and pathophysiological consequences of abdominal adiposity and abdominal adipose tissue depots. Nutrition. 2003;19: Madan K, Bhardwaj P, Thareja S, Gupta SD, Saraya A. Oxidant stress and antioxidant status among patients with nonalcoholic fatty liver disease (NAFLD). J Clin Gastroenterol. 2006;40: Bhardwaj P, Madan K, Thareja S, Joshi YK, Saraya A. Comparative redox status in alcoholic liver disease and nonalcoholic fatty liver disease. Hepatol Int. 2008;2: Narasimhan S, Gokulakrishnan K, Sampathkumar R, et al. Oxidative stress is independently associated with non-alcoholic fatty liver disease (NAFLD) in subjects with and without type 2 diabetes. Clin Biochem. 2010;43: Das KS, Balakrishnan V, Mukherjee S, Vasudevan DM. Evaluation of blood oxidative stress-related parameters in alcoholic liver disease and non-alcoholic fatty liver disease. Scand J Clin Lab Invest. 2008;68: Duseja A, Das R, Das A, Dhiman RK, Chawla Y, Garewal G. Serum iron and hepatic iron overload in patients with nonalcoholic steatohepatitis. Dig Dis Sci. 2006;51: Duseja A, Das R, Nanda M, Das A, Garewal G, Chawla Y. Nonalcoholic steatohepatitis in Asian Indians is neither associated with iron overload nor with HFE gene mutations. World J Gastroenterol. 2005;21:393 5.

9 Indian J Gastroenterol (2010) 29: Sharma R, Sinha S, Danishad KA, et al. Investigation of hepatic gluconeogenesis pathway in non-diabetic Asian Indians with nonalcoholic fatty liver disease using in vivo ((31)P) phosphorus magnetic resonance spectroscopy. Atherosclerosis. 2009;203: Nagarajan P, Venkatesan R, Kumar M, Usmani A, Majumdar SS. Macaca radiata (bonnet monkey): a spontaneous model of nonalcoholic fatty liver disease. Liver Int. 2008;28: Duseja A, Das A, Das R, Dhiman RK, Chawla Y, Bhansali A. Unconjugated hyperbilirubinemia in nonalcoholic steatohepatitis is it Gilbert s syndrome? Trop Gastroenterol. 2005;26: Kalra N, Duseja A, Das A, et al. Chemical shift magnetic resonance imaging is helpful in detecting hepatic steatosis but not fibrosisin patients with nonalcoholic fatty liver disease (NAFLD). Ann Hepatol. 2009;8: Vasdev N, Kakati AG, Saigal S, Nayak NC. Spectrum of histological features in non-alcoholic fatty liver disease. Natl Med J India. 2007;20: Singh DK, Sakhuja P, Malhotra V, Gondal R, Sarin SK. Independent predictors of steatohepatitis and fibrosis in Asian Indian patients with non-alcoholic steatohepatitis. Dig Dis Sci. 2008;53: Musso G, Gambino R, Cassader M, Pagano G. A meta-analysis of randomized trials for the treatment of nonalcoholic fatty liver disease. Hepatology. 2010;52: Duseja A, Das A, Dhiman RK, et al. Metformin is effective in achieving biochemical response in patients with nonalcoholic fatty liver disease (NAFLD) not responding to lifestyle interventions. Ann Hepatol. 2007;6: Madan K, Batra Y, Gupta DS, et al. Vitamin E-based therapy is effective in ameliorating transaminasemia in nonalcoholic fatty liver disease. Indian J Gastroenterol. 2005;24: Satapathy SK, Garg S, Chauhan R, et al. Beneficial affects of tumor necrosis factor-alfa inhibition by pentoxifylline on clinical, biochemical, and metabolic parameters of patients with nonalcoholic steatohepatitis. Am J Gastroenterol. 2004;99: Satapathy SK, Sakhuja P, Malhotra V, Sharma BC, Sarin SK. Beneficial effects of pentoxifylline on hepatic steatosis, fibrosis and necroinflammation in patients with non-alcoholic steatohepatitis. J Gastroenterol Hepatol. 2007;22: Sanal MG. Adipose tissue transplantation may be a potential treatment for diabetes, atherosclerosis and nonalcoholic steatohepatitis. Med Hypotheses. 2009;72: Petersen KF, Dufour S, Feng J, et al. Increased prevalence of insulin resistance and nonalcoholic fatty liver disease in Asian- Indian men. Proc Natl Acad Sci USA. 2006;103: Duseja A. Nonalcoholic fatty liver disease in India is it different? Trop Gastroenterol. 2006;27: Amarapurkar DN, Hashimoto E, Lesmana LA, Sollano JD, Chen PJ. Asia-Pacific Working Party on NAFLD. How common is nonalcoholic fatty liver disease in the Asia-Pacific region and are there local differences? J Gastroenterol Hepatol. 2007;22: