NASH: A Case for Personalized Treatment
NASH: A Case for Personalized Treatment
Past studies investigating lifestyle modifications (diet, exercise, behaviour modification) found that programmes that lead to a significant weight loss (5–10% of total body weight) improved steatosis. Greater weight loss of at least 10% has also been shown to improve inflammation. Recently, Eckard and group recently completed a lifestyle modification study, noting the impact exercise and diet had on histopathological changes. A total of 56 participants, broken into four groups based on different diet components and length of exercise regimens, were included in the study. They found that many of the lifestyle modifications made a significant difference to the histopathological profile further studies with larger group sizes will be needed to determine any difference between the subgroups. However, it is important to note that adherence to lifestyle interventions can be problematic. In several studies, only 15% of patients achieved a weight loss ≥10% and most of them regained their weight, with the best average weight loss among participants being 3–4 kg at 2 and 4 years.
On the other hand, studies that investigated exercise only without dietary modification reported that exercise alone reduced hepatic fat, but had variable effect on hepatic inflammation. However, there are potential barriers to exercise. One study found that, although NASH patients knew that exercise was important, they lacked confidence in the performance of the exercises, citing things such as afraid of falling, which was directly proportional to the degree of difficulty performing the exercise.
In addition, another factor that impairs a patient with NASH in performing exercise is fatigue, one of the most prevalent and pervasive symptoms in people with NASH. A multivariate analysis showed that increased cognitive difficulty, age, fatigue, lower albumin and bilirubin were all independently associated with functional difficulty in patients with NASH. Fatigue has been shown to interfere with daily routines, including work, school, social, household and leisure activities. These potential contributors should be evaluated prior to prescribing an exercise plan because they are likely to influence the exercise prescription and contribute to whether an individual will respond to exercise. With these factors in mind, there are no recommended guidelines for an exercise programme specifically designed for patients with NASH. However, many practitioners follow the American College of Sports Medicine's protocols for setting exercise goals and establishing a starting point.
The second part of 'lifestyle modification' is dietary changes. In their recent meta- analysis, Musso and colleagues determined that caloric restriction is the most important goal for patients with NASH to improve hepatic steatosis, but depending on the patient, diets of different composition may bring even greater benefits, i.e. diabetic patients had improved hepatic insulin sensitivity when placed on a low-carbohydrate diet and for those with cholesterol issues, a low-fat diet served them better by improving the LDL-C and HDL-C.
Other investigators determined that avoiding drinks and food high in fructose and trans fats is essential. Fructose and trans fats are associated with insulin resistance and the development of hepatic steatosis is thought to occur as a consequence ATP depletion, lipotoxicity and tumour necrosis factor (TNF) expression. Therefore, current recommendations for patients with NASH are to avoid drinks with fructose and foods high in trans fat and seek out foods high in polyunsaturated fats.
Finally, ingestion of coffee may be very beneficial for patients with NASH as well. Although some findings of coffee consumption are contradictory, overall, drinking coffee (but not expresso) has been found to have a protective effect in NASH as well as in biopsy-proven NASH through the reduction of hepatic inflammation and fibrosis. The exact mechanism of how coffee exerts its anti-inflammatory/fibrotic effect is still under investigation, but investigators now believe another mechanism related to UDP glucuronosyltransferases may also be involved.
In summary, the current evidence suggests that significant weight loss and possibly dietary change may positively impact surrogate markers of NASH. Nevertheless, the impact on the most important clinical outcomes is not available. On the other hand, data showing the efficacy of exercise in modifying outcomes of NASH are even less compelling. In 2013, weight loss, dietary changes and exercise should be recommended for all patients with NASH, primarily to modify their cardiovascular risks rather than liver-related risks.
Orlistat is a drug that inhibits enteric lipid absorption and has been promoted as a weight-loss aid. Orlistat, in addition to lifestyle modification, was evaluated as a potential therapy for NASH. Although ALT levels were significantly decreased in both groups, the orlistat group showed a larger reduction. There was also a reduction in fatty liver by ultrasound in the orlistat group. However, this finding could not be confirmed by biopsy due to lack of follow-up. In another trial, calorie-restricted patients were asked to take vitamin E and then were randomly assigned to either a placebo group or a group that was administered orlistat. Patients taking orlistat did not benefit from significant weight loss or histological improvement compared with the placebo group. If there was any histological improvement, the improvement was proportional to weight loss and not attributed specifically to orlistat.
Harrison and group conducted a randomised control study to investigate the effects of orlistat. They determined that orlistat was safe and well tolerated; however, the ingestion of orlistat did not add any further cardio-metabolic or histological changes over lifestyle modification alone. Based on these data, the evidence supporting the efficacy of orlistat in patients with NASH is lacking, hence this regimen cannot be highly recommended as a potential therapy for patients with NASH.
Metformin. Metformin, a drug well known for its effect on insulin resistance, has also been investigated regarding its use in patients with NASH. Although many of the studies showed an improvement in liver enzymes while taking metformin, not all studies have reported the same results. Improvement in histology has also been evaluated with variable results. One small trial, looking at patients with NASH, concluded that the modest decrease in ALT levels and histology probably reflected effects from weight loss in general and not effect of metformin. Another small, open-label trial noted a decrease in ALT and AST; however, this was not sustainable and, after 3 months, the liver enzymes increased. In addition, there was only a minimal improvement in histology. Metformin was also investigated in a trial that placed nondiabetic NASH patients on a diet and exercise programme vs. metformin patients on a diet and exercise programme plus a placebo. The results of this study demonstrated that the metformin had little effect on the liver histology or on aminotransferases. Furthermore, another small randomised clinical trial placed patients on metformin for 6 months and found no improvement in histology. Finally, a recent randomised study of metformin, the TONIC trial, compared metformin with vitamin E and placebo in children with NASH. Again, there was no significant reduction in ALT compared with placebo in the metformin group and, with the exception of hepatocellular ballooning, there was no significant change in histology. Based on these study results and others, the current NAFLD guidelines do not recommend metformin as a treatment for NAFLD. Although metformin can be used for diabetic patients with NASH, there is no evidence to support its efficacy solely from the standpoint of NASH.
Thiazolidinediones. The majority of the studies examining thiazolidinediones (TZDs), medications known to enhance insulin sensitivity, which prevents the activation of adiponectin thus impairing adipocyte storage of triglycerides, involve pioglitazone and rosiglitazone. Rosiglitazone has been shown to decrease aminotransferases and improve histological inflammation, but not fibrosis. Subsequent studies have confirmed the improvement in the liver enzymes, but did not show histological improvement.
In addition to rosiglitazone, pioglitazone has also been investigated for treatment of NASH. A recent randomised clinical trial assessed the efficacy of pioglitazone vs. a placebo in patients with NASH. In this study, patients receiving pioglitazone showed improved histology and fibrosis. On the other hand, the PIVENS trial randomly assigned 247 nondiabetic patients with NASH to one of three groups: pioglitazone, vitamin E or placebo group. Liver biopsy results were assessed after 96 weeks of therapy. Although improvement was noted in both the vitamin E and pioglitazone groups, the rate of improvement in the pioglitazone did not meet the predetermined histological criteria. It is important to note that both Vitamin E and pioglitazone were associated with a decrease in liver enzymes as well as with improvement in steatosis and lobular inflammation. Although histological improvement was noted in the pioglitazone group, the improvement in the vitamin E group was greater, 34% vs. 43%, reaching the primary end point of histological improvement respectively. Nevertheless, neither group showed improvement in hepatic fibrosis. Of note, the patients in the pioglitazone group gained more weight than those taking placebo or vitamin E. A limitation of the PIVENS study may have been related to the use of pioglitazone in nondiabetic NASH patients whose underlying liver disease may not have been primarily related to this pathway. Another limitation of PIVENS was that some of these patients did not have true NASH, therefore underpowering the study. It is possible that the predominant pathogenic pathway for this cohort may have been related to oxidative stress, favouring vitamin E as antioxidant. Although vitamin E can be recommended as a first-line treatment for NASH patients without diabetes, the long-term safety remains an issue (see next section). On the other hand, safety of pioglitazone in nondiabetics with NASH has not been established as TZDs have been associated with weight gain (average is 8.8 lbs), congestive heart failure, bone loss with increased risk of fractures and an increase in bladder cancers. In addition, the durability of response on TZD therapy is questionable, as Lutchman and colleagues noted in their study on the effects of discontinuing pioglitazone in patients with NASH. They noted that, after the discontinuation of TZD, there was an increase in aminotransferases, an increase in HOMA after 1 year and a recurrence of steatohepatitis. These findings have called into question the duration of therapy. Therefore, careful consideration must be given to the risks and benefits for each patient before administering a TZD as a treatment option for diabetic patients with NASH.
Vitamin E. Vitamin E is thought to decrease inflammation by acting as an antioxidant, thus decreasing oxidative cell damage. Although a number of open-label clinical trials have suggested the potential efficacy of vitamin E, the best evidence comes from the PIVENS trial. As noted previously, the PIVENS trial showed a greater histological improvement in inflammation, but not in fibrosis, in nondiabetic patients taking vitamin E compared with those taking placebo and those taking pioglitazone. Although a larger proportion of patients showed histological improvement in the vitamin E group, it is important to note that less than half of all the patients had any improvement. Another smaller randomised trial published a year later, dubbed the TONIC trial, compared vitamin E with metformin in paediatric patients with biopsy-proven NASH. The primary outcome was reduction in alanine aminotransferase (ALT), the secondary outcome was improvement in histology and resolution of NASH. Compared with placebo, neither intervention demonstrated significant, sustained reductions in ALT levels. Of note, initially, there was a greater reduction in ALT levels in those children taking Vitamin E; however, the findings were not sustained over time. Interestingly, in the children taking vitamin E, there was a significant improvement noted in hepatocellular ballooning, NAFLD activity score and resolution of NASH in patients with borderline NASH. It is important to mention that there has been some concern with the long-term use of Vitamin E. In fact, a meta-analysis has reported an increase in all-cause mortality in patients using vitamin E for cardiovascular conditions. The recent NAFLD guideline recommends Vitamin E as a first-line treatment for nondiabetic patients with only biopsy-proven NASH. On the other hand, Vitamin E is not recommended for NASH patients with diabetes due to lack of data on long-term safety and efficacy.
Ursodeoxycholic Acid (UDCA). Ursodeoxycholic acid therapy decreases the cholesterol secretion into bile, as indicated by a decline in the cholesterol fraction of biliary lipids. UDCA exerts its action(s) in liver through multiple possibly interrelated pathways, including alterations of bile acid pool, choleresis, immune modulation effects and cytoprotection mechanism. Although UDCA was initially suggested as a promising drug for the treatment of NASH, literature does not support its use. In fact, in a randomised controlled trial, 166 NASH patients were randomised to UDCA vs. placebo for 2 years. Of 166 patients enrolled, 126 patients completed the study; however, only 107 patients had pre- and post-treatment liver biopsies available. The biopsies available showed no significant difference in histology between the two groups. Only the study by Dufour et al. observed a significant improvement of NASH with the combination of UDCA and Vitamin E when compared with a placebo therapy. The administration of only UDCA was not as effective in the treatment of NASH, although there was much heterogeneity among the three other studies. Therefore, at the current time, evidence to support the use of UDCA is lacking, and its use cannot be recommended.
Statins. Although there are number of open-label trials suggesting improvement in liver enzymes, no convincing histological data are available. Therefore, statins are not currently recommended to treat patients with NASH for their liver disease. On the other hand, statins are safe to use for the purpose of treating hyperlipidaemia in these individuals who also have cardiovascular disease, and the primary indication for use of statins is as an anti-cholesterol agent. Therefore, clinicians treating hyperlipidaemia in NASH patients should not rule out the use of statins in treating hyperlipidaemia and should be considered as part of their therapeutic armamentarium.
Other Treatment Regimens. Other regimens such as omega-3 fatty acids, pentoxifylline, probiotics and traditional Chinese medicine such as Hawthorn fruit have been studied for treatment of NASH. Parker and colleagues conducted a meta-analysis on the use of omega 3 polyunsaturated fatty acid (PUFA) on NASH. The primary outcome measures of interest were liver fat and liver function tests: alanine aminotransferase (ALT) and aspartate aminotransferase (AST). There were nine eligible studies involving 355 individuals in the meta- analysis. There were beneficial changes in liver fat for those receiving PUFA treatment as well as an observed reduction in AST and a trend towards favouring PUFA treatment on ALT. Despite the heterogeneity of the studies, the pooled data suggested that omega-3 PUFA supplementation may decrease liver fat; however, the optimal dose is not currently known.
Pentoxifylline is considered as anti-TNF compound. It decreases the inflammation associated with NASH. Although one study suggested improvement in steatosis, improvement in other important laboratory and histological outcomes was lacking.
Probiotics, in theory, offer an interesting approach to NASH. Gut flora is hypothesised to play a role in energy regulation and therefore adiposity. Also, endotoxins from the gut flora can get to the liver through portal blood system and could represent a second hit. In fact, probiotics have been suggested to be beneficial in NASH by augmenting the microbial environment in the intestine. In addition, they may provide anti-inflammatory protection. Despite early studies, no well-designed clinical trial data are available to provide evidence to support the use of any of these regimens for patients with NASH.
Shi and colleagues conducted a meta-analysis on the use of traditional Chinese herbal medicines (TCM's) compared with the western medications, including placebo, ursodeoxycholic acid, insulin sensitisers, lipid-lowering drugs or antioxidants. Five thousand nine hundred and four patients from 62 RCTs were included for the meta-analysis and 25 661 patients from 419 clinical studies were included for TCM formulation analysis. These investigators found that, when compared with western medicines, TCMs had a better effect on the normalisation of alanine aminotransferase and the disappearance of radiological steatosis in the treatment of NAFLD. There were 246 kinds of Chinese herbs that were included in the study, with an average of 10 herbs (range 1–31) in each formulation. Hawthorn Fruit (321 times in 17 670 patients) was the most often used herb in the treatment of NAFLD. They concluded that TCM is of modest benefit in the treatment of NAFLD.
Although there are some promising results with these therapeutic medications, without further well-controlled randomised studies, omega-3 fatty acids, pentoxifylline, probiotics and traditional Chinese medicine such as Hawthorn fruit cannot be recommended as first-line therapeutic agents to treat NASH. It is also important to emphasize that consideration of safety of herbal supplements is critical and must be carefully assessed.
Bariatric Surgery. The impact that bariatric surgery has on NAFLD has been assessed with liver enzymes and histology. In a number of studies, aminotransferases and GGT levels, as surrogate markers of liver histology, have shown some improvement. One study found improvement in histology and fibrosis score 1 year post-operatively, whereas others have shown improvement in histology, but not necessarily in fibrosis. A meta-analysis reviewed 15 studies and compiled results for improvement in steatosis, steatohepatitis and fibrosis as well as a mean reduction in body mass index. The study concluded that steatosis, steatohepatitis and fibrosis all improved and, in some cases, resolved after weight loss associated with bariatric surgery. A prospective trial performed in 2009 followed severely obese patients for 5 years post-bariatric surgery. The overall percentage of biopsies of patients with steatosis was lower at 5 years. Conversely, 5 years after surgery, the level of fibrosis increased significantly. Levels of insulin resistance were independently predictive of steatosis and ballooning at the 5-year mark. A randomised, nonblinded clinical trial was published in 2012 comparing bariatric surgery with intensive medical therapy in obese patients with diabetes. Although the study did not include liver histology as primary or secondary end point, bariatric surgery was more effective at controlling glucose levels than medical therapy alone in these patients.
As diabetes and obesity are strong risk factors for the development of NASH, one can assume that better regimen for sustained weight loss and glycaemic control could potentially improve NASH outcomes. In fact, recent studies investigating the surgical management of patients with morbid obesity and NASH found that weight loss following conventional nonmalabsorptive bariatric-metabolic procedures reduces steatosis and lobular inflammation, but may not have a consistent effect on liver fibrosis. However, rapid weight loss, especially with malabsorptive procedures, may actually produce a transient or prolonged increase in liver disease, and therefore; patients need careful monitoring to make sure their weight loss occurs over time. It can be concluded that bariatric-metabolic surgery in patients with compensated cirrhosis is not established, but is definitely contraindicated in decompensated cirrhosis.
Given the lack of strong published evidence, the NAFLD guideline does not formally recommend bariatric surgery for treatment of NASH. On the other hand, it is important to note that bariatric surgery is not contraindicated in NASH patients without cirrhosis. Additional well-designed studies are needed to assess the efficacy of this important treatment modality for obese-diabetic patients with NASH.
Specific Strategies to Treat NASH
Weight Loss and Lifestyle Modification
Past studies investigating lifestyle modifications (diet, exercise, behaviour modification) found that programmes that lead to a significant weight loss (5–10% of total body weight) improved steatosis. Greater weight loss of at least 10% has also been shown to improve inflammation. Recently, Eckard and group recently completed a lifestyle modification study, noting the impact exercise and diet had on histopathological changes. A total of 56 participants, broken into four groups based on different diet components and length of exercise regimens, were included in the study. They found that many of the lifestyle modifications made a significant difference to the histopathological profile further studies with larger group sizes will be needed to determine any difference between the subgroups. However, it is important to note that adherence to lifestyle interventions can be problematic. In several studies, only 15% of patients achieved a weight loss ≥10% and most of them regained their weight, with the best average weight loss among participants being 3–4 kg at 2 and 4 years.
On the other hand, studies that investigated exercise only without dietary modification reported that exercise alone reduced hepatic fat, but had variable effect on hepatic inflammation. However, there are potential barriers to exercise. One study found that, although NASH patients knew that exercise was important, they lacked confidence in the performance of the exercises, citing things such as afraid of falling, which was directly proportional to the degree of difficulty performing the exercise.
In addition, another factor that impairs a patient with NASH in performing exercise is fatigue, one of the most prevalent and pervasive symptoms in people with NASH. A multivariate analysis showed that increased cognitive difficulty, age, fatigue, lower albumin and bilirubin were all independently associated with functional difficulty in patients with NASH. Fatigue has been shown to interfere with daily routines, including work, school, social, household and leisure activities. These potential contributors should be evaluated prior to prescribing an exercise plan because they are likely to influence the exercise prescription and contribute to whether an individual will respond to exercise. With these factors in mind, there are no recommended guidelines for an exercise programme specifically designed for patients with NASH. However, many practitioners follow the American College of Sports Medicine's protocols for setting exercise goals and establishing a starting point.
The second part of 'lifestyle modification' is dietary changes. In their recent meta- analysis, Musso and colleagues determined that caloric restriction is the most important goal for patients with NASH to improve hepatic steatosis, but depending on the patient, diets of different composition may bring even greater benefits, i.e. diabetic patients had improved hepatic insulin sensitivity when placed on a low-carbohydrate diet and for those with cholesterol issues, a low-fat diet served them better by improving the LDL-C and HDL-C.
Other investigators determined that avoiding drinks and food high in fructose and trans fats is essential. Fructose and trans fats are associated with insulin resistance and the development of hepatic steatosis is thought to occur as a consequence ATP depletion, lipotoxicity and tumour necrosis factor (TNF) expression. Therefore, current recommendations for patients with NASH are to avoid drinks with fructose and foods high in trans fat and seek out foods high in polyunsaturated fats.
Finally, ingestion of coffee may be very beneficial for patients with NASH as well. Although some findings of coffee consumption are contradictory, overall, drinking coffee (but not expresso) has been found to have a protective effect in NASH as well as in biopsy-proven NASH through the reduction of hepatic inflammation and fibrosis. The exact mechanism of how coffee exerts its anti-inflammatory/fibrotic effect is still under investigation, but investigators now believe another mechanism related to UDP glucuronosyltransferases may also be involved.
In summary, the current evidence suggests that significant weight loss and possibly dietary change may positively impact surrogate markers of NASH. Nevertheless, the impact on the most important clinical outcomes is not available. On the other hand, data showing the efficacy of exercise in modifying outcomes of NASH are even less compelling. In 2013, weight loss, dietary changes and exercise should be recommended for all patients with NASH, primarily to modify their cardiovascular risks rather than liver-related risks.
Orlistat for Treatment of NASH
Orlistat is a drug that inhibits enteric lipid absorption and has been promoted as a weight-loss aid. Orlistat, in addition to lifestyle modification, was evaluated as a potential therapy for NASH. Although ALT levels were significantly decreased in both groups, the orlistat group showed a larger reduction. There was also a reduction in fatty liver by ultrasound in the orlistat group. However, this finding could not be confirmed by biopsy due to lack of follow-up. In another trial, calorie-restricted patients were asked to take vitamin E and then were randomly assigned to either a placebo group or a group that was administered orlistat. Patients taking orlistat did not benefit from significant weight loss or histological improvement compared with the placebo group. If there was any histological improvement, the improvement was proportional to weight loss and not attributed specifically to orlistat.
Harrison and group conducted a randomised control study to investigate the effects of orlistat. They determined that orlistat was safe and well tolerated; however, the ingestion of orlistat did not add any further cardio-metabolic or histological changes over lifestyle modification alone. Based on these data, the evidence supporting the efficacy of orlistat in patients with NASH is lacking, hence this regimen cannot be highly recommended as a potential therapy for patients with NASH.
Treatment of Insulin Resistance Associated With NASH and Liver Fibrosis
Metformin. Metformin, a drug well known for its effect on insulin resistance, has also been investigated regarding its use in patients with NASH. Although many of the studies showed an improvement in liver enzymes while taking metformin, not all studies have reported the same results. Improvement in histology has also been evaluated with variable results. One small trial, looking at patients with NASH, concluded that the modest decrease in ALT levels and histology probably reflected effects from weight loss in general and not effect of metformin. Another small, open-label trial noted a decrease in ALT and AST; however, this was not sustainable and, after 3 months, the liver enzymes increased. In addition, there was only a minimal improvement in histology. Metformin was also investigated in a trial that placed nondiabetic NASH patients on a diet and exercise programme vs. metformin patients on a diet and exercise programme plus a placebo. The results of this study demonstrated that the metformin had little effect on the liver histology or on aminotransferases. Furthermore, another small randomised clinical trial placed patients on metformin for 6 months and found no improvement in histology. Finally, a recent randomised study of metformin, the TONIC trial, compared metformin with vitamin E and placebo in children with NASH. Again, there was no significant reduction in ALT compared with placebo in the metformin group and, with the exception of hepatocellular ballooning, there was no significant change in histology. Based on these study results and others, the current NAFLD guidelines do not recommend metformin as a treatment for NAFLD. Although metformin can be used for diabetic patients with NASH, there is no evidence to support its efficacy solely from the standpoint of NASH.
Thiazolidinediones. The majority of the studies examining thiazolidinediones (TZDs), medications known to enhance insulin sensitivity, which prevents the activation of adiponectin thus impairing adipocyte storage of triglycerides, involve pioglitazone and rosiglitazone. Rosiglitazone has been shown to decrease aminotransferases and improve histological inflammation, but not fibrosis. Subsequent studies have confirmed the improvement in the liver enzymes, but did not show histological improvement.
In addition to rosiglitazone, pioglitazone has also been investigated for treatment of NASH. A recent randomised clinical trial assessed the efficacy of pioglitazone vs. a placebo in patients with NASH. In this study, patients receiving pioglitazone showed improved histology and fibrosis. On the other hand, the PIVENS trial randomly assigned 247 nondiabetic patients with NASH to one of three groups: pioglitazone, vitamin E or placebo group. Liver biopsy results were assessed after 96 weeks of therapy. Although improvement was noted in both the vitamin E and pioglitazone groups, the rate of improvement in the pioglitazone did not meet the predetermined histological criteria. It is important to note that both Vitamin E and pioglitazone were associated with a decrease in liver enzymes as well as with improvement in steatosis and lobular inflammation. Although histological improvement was noted in the pioglitazone group, the improvement in the vitamin E group was greater, 34% vs. 43%, reaching the primary end point of histological improvement respectively. Nevertheless, neither group showed improvement in hepatic fibrosis. Of note, the patients in the pioglitazone group gained more weight than those taking placebo or vitamin E. A limitation of the PIVENS study may have been related to the use of pioglitazone in nondiabetic NASH patients whose underlying liver disease may not have been primarily related to this pathway. Another limitation of PIVENS was that some of these patients did not have true NASH, therefore underpowering the study. It is possible that the predominant pathogenic pathway for this cohort may have been related to oxidative stress, favouring vitamin E as antioxidant. Although vitamin E can be recommended as a first-line treatment for NASH patients without diabetes, the long-term safety remains an issue (see next section). On the other hand, safety of pioglitazone in nondiabetics with NASH has not been established as TZDs have been associated with weight gain (average is 8.8 lbs), congestive heart failure, bone loss with increased risk of fractures and an increase in bladder cancers. In addition, the durability of response on TZD therapy is questionable, as Lutchman and colleagues noted in their study on the effects of discontinuing pioglitazone in patients with NASH. They noted that, after the discontinuation of TZD, there was an increase in aminotransferases, an increase in HOMA after 1 year and a recurrence of steatohepatitis. These findings have called into question the duration of therapy. Therefore, careful consideration must be given to the risks and benefits for each patient before administering a TZD as a treatment option for diabetic patients with NASH.
Antioxidants and Cytoprotective Agents
Vitamin E. Vitamin E is thought to decrease inflammation by acting as an antioxidant, thus decreasing oxidative cell damage. Although a number of open-label clinical trials have suggested the potential efficacy of vitamin E, the best evidence comes from the PIVENS trial. As noted previously, the PIVENS trial showed a greater histological improvement in inflammation, but not in fibrosis, in nondiabetic patients taking vitamin E compared with those taking placebo and those taking pioglitazone. Although a larger proportion of patients showed histological improvement in the vitamin E group, it is important to note that less than half of all the patients had any improvement. Another smaller randomised trial published a year later, dubbed the TONIC trial, compared vitamin E with metformin in paediatric patients with biopsy-proven NASH. The primary outcome was reduction in alanine aminotransferase (ALT), the secondary outcome was improvement in histology and resolution of NASH. Compared with placebo, neither intervention demonstrated significant, sustained reductions in ALT levels. Of note, initially, there was a greater reduction in ALT levels in those children taking Vitamin E; however, the findings were not sustained over time. Interestingly, in the children taking vitamin E, there was a significant improvement noted in hepatocellular ballooning, NAFLD activity score and resolution of NASH in patients with borderline NASH. It is important to mention that there has been some concern with the long-term use of Vitamin E. In fact, a meta-analysis has reported an increase in all-cause mortality in patients using vitamin E for cardiovascular conditions. The recent NAFLD guideline recommends Vitamin E as a first-line treatment for nondiabetic patients with only biopsy-proven NASH. On the other hand, Vitamin E is not recommended for NASH patients with diabetes due to lack of data on long-term safety and efficacy.
Ursodeoxycholic Acid (UDCA). Ursodeoxycholic acid therapy decreases the cholesterol secretion into bile, as indicated by a decline in the cholesterol fraction of biliary lipids. UDCA exerts its action(s) in liver through multiple possibly interrelated pathways, including alterations of bile acid pool, choleresis, immune modulation effects and cytoprotection mechanism. Although UDCA was initially suggested as a promising drug for the treatment of NASH, literature does not support its use. In fact, in a randomised controlled trial, 166 NASH patients were randomised to UDCA vs. placebo for 2 years. Of 166 patients enrolled, 126 patients completed the study; however, only 107 patients had pre- and post-treatment liver biopsies available. The biopsies available showed no significant difference in histology between the two groups. Only the study by Dufour et al. observed a significant improvement of NASH with the combination of UDCA and Vitamin E when compared with a placebo therapy. The administration of only UDCA was not as effective in the treatment of NASH, although there was much heterogeneity among the three other studies. Therefore, at the current time, evidence to support the use of UDCA is lacking, and its use cannot be recommended.
Statins. Although there are number of open-label trials suggesting improvement in liver enzymes, no convincing histological data are available. Therefore, statins are not currently recommended to treat patients with NASH for their liver disease. On the other hand, statins are safe to use for the purpose of treating hyperlipidaemia in these individuals who also have cardiovascular disease, and the primary indication for use of statins is as an anti-cholesterol agent. Therefore, clinicians treating hyperlipidaemia in NASH patients should not rule out the use of statins in treating hyperlipidaemia and should be considered as part of their therapeutic armamentarium.
Other Treatment Regimens. Other regimens such as omega-3 fatty acids, pentoxifylline, probiotics and traditional Chinese medicine such as Hawthorn fruit have been studied for treatment of NASH. Parker and colleagues conducted a meta-analysis on the use of omega 3 polyunsaturated fatty acid (PUFA) on NASH. The primary outcome measures of interest were liver fat and liver function tests: alanine aminotransferase (ALT) and aspartate aminotransferase (AST). There were nine eligible studies involving 355 individuals in the meta- analysis. There were beneficial changes in liver fat for those receiving PUFA treatment as well as an observed reduction in AST and a trend towards favouring PUFA treatment on ALT. Despite the heterogeneity of the studies, the pooled data suggested that omega-3 PUFA supplementation may decrease liver fat; however, the optimal dose is not currently known.
Pentoxifylline is considered as anti-TNF compound. It decreases the inflammation associated with NASH. Although one study suggested improvement in steatosis, improvement in other important laboratory and histological outcomes was lacking.
Probiotics, in theory, offer an interesting approach to NASH. Gut flora is hypothesised to play a role in energy regulation and therefore adiposity. Also, endotoxins from the gut flora can get to the liver through portal blood system and could represent a second hit. In fact, probiotics have been suggested to be beneficial in NASH by augmenting the microbial environment in the intestine. In addition, they may provide anti-inflammatory protection. Despite early studies, no well-designed clinical trial data are available to provide evidence to support the use of any of these regimens for patients with NASH.
Shi and colleagues conducted a meta-analysis on the use of traditional Chinese herbal medicines (TCM's) compared with the western medications, including placebo, ursodeoxycholic acid, insulin sensitisers, lipid-lowering drugs or antioxidants. Five thousand nine hundred and four patients from 62 RCTs were included for the meta-analysis and 25 661 patients from 419 clinical studies were included for TCM formulation analysis. These investigators found that, when compared with western medicines, TCMs had a better effect on the normalisation of alanine aminotransferase and the disappearance of radiological steatosis in the treatment of NAFLD. There were 246 kinds of Chinese herbs that were included in the study, with an average of 10 herbs (range 1–31) in each formulation. Hawthorn Fruit (321 times in 17 670 patients) was the most often used herb in the treatment of NAFLD. They concluded that TCM is of modest benefit in the treatment of NAFLD.
Although there are some promising results with these therapeutic medications, without further well-controlled randomised studies, omega-3 fatty acids, pentoxifylline, probiotics and traditional Chinese medicine such as Hawthorn fruit cannot be recommended as first-line therapeutic agents to treat NASH. It is also important to emphasize that consideration of safety of herbal supplements is critical and must be carefully assessed.
Bariatric Surgery. The impact that bariatric surgery has on NAFLD has been assessed with liver enzymes and histology. In a number of studies, aminotransferases and GGT levels, as surrogate markers of liver histology, have shown some improvement. One study found improvement in histology and fibrosis score 1 year post-operatively, whereas others have shown improvement in histology, but not necessarily in fibrosis. A meta-analysis reviewed 15 studies and compiled results for improvement in steatosis, steatohepatitis and fibrosis as well as a mean reduction in body mass index. The study concluded that steatosis, steatohepatitis and fibrosis all improved and, in some cases, resolved after weight loss associated with bariatric surgery. A prospective trial performed in 2009 followed severely obese patients for 5 years post-bariatric surgery. The overall percentage of biopsies of patients with steatosis was lower at 5 years. Conversely, 5 years after surgery, the level of fibrosis increased significantly. Levels of insulin resistance were independently predictive of steatosis and ballooning at the 5-year mark. A randomised, nonblinded clinical trial was published in 2012 comparing bariatric surgery with intensive medical therapy in obese patients with diabetes. Although the study did not include liver histology as primary or secondary end point, bariatric surgery was more effective at controlling glucose levels than medical therapy alone in these patients.
As diabetes and obesity are strong risk factors for the development of NASH, one can assume that better regimen for sustained weight loss and glycaemic control could potentially improve NASH outcomes. In fact, recent studies investigating the surgical management of patients with morbid obesity and NASH found that weight loss following conventional nonmalabsorptive bariatric-metabolic procedures reduces steatosis and lobular inflammation, but may not have a consistent effect on liver fibrosis. However, rapid weight loss, especially with malabsorptive procedures, may actually produce a transient or prolonged increase in liver disease, and therefore; patients need careful monitoring to make sure their weight loss occurs over time. It can be concluded that bariatric-metabolic surgery in patients with compensated cirrhosis is not established, but is definitely contraindicated in decompensated cirrhosis.
Given the lack of strong published evidence, the NAFLD guideline does not formally recommend bariatric surgery for treatment of NASH. On the other hand, it is important to note that bariatric surgery is not contraindicated in NASH patients without cirrhosis. Additional well-designed studies are needed to assess the efficacy of this important treatment modality for obese-diabetic patients with NASH.
