| |
Fatty Liver Accelerates Fibrosis: diabetes, mitochondrial dyfunction, obesity associated
|
| |
| |
"Natural history of nonalcoholic steatohepatitis: A longitudinal study of repeat liver biopsies"
Hepatology
Volume 40, Issue 4, October 2004
Eduardo Fassio 1 *, Estela Álvarez 2, Nora Domínguez 1, Graciela Landeira 1, Cristina Longo 1
1Hepatology Unit, Gastroenterology Service, Hospital Nacional Profesor Alejandro Posadas, El Palomar, Buenos Aires, Argentina
2Pathology Service, Hospital Nacional Profesor Alejandro Posadas, El Palomar, Buenos Aires, Argentina
"...Our study shows that 31.8% of 22 patients with NASH had a progression of liver fibrosis over a median follow-up of 4.3 years. These findings confirmed the progressive potential of NASH. However, most of the patients (68.2%) had no fibrosis progression...
... A two-hit hypothesis has been proposed to explain the mechanisms of liver damage in NASH... The first hit has been associated with insulin resistance... in only a proportion of patients with insulin resistance and NASH will an advanced liver disease develop, and other factors, or hits (like oxidative stress, mitochondrial dysfunction, and abnormal cytokine production), should be present to produce severe cellular injury or fibrosis...
... Among 12 variables studied at baseline, we found that the prevalence of obesity and BMI were the only variables [associated with fibrosis]... Presence of diabetes has been shown to be an independent predictor of severe fibrosis in a retrospective study that analyzed a large number of patients. Among our patients, four of seven (57%) from group P and 4 of 15 (27%) from group NP had diabetes...
... Some very recent, noncontrolled studies have shown that insulin-sensitizing agents (rosiglitazone and pioglitazone) can lead to improvement in histological features in NASH patients after 48 weeks of treatment. If these results were confirmed in larger, controlled studies, this kind of therapy would be accepted as being able to modify the natural evolution of NASH and would be recommended promptly for clinical practice..."
ABSTRACT/SUMMARY
Nonalcoholic steatohepatitis may cause severe fibrosis, cirrhosis, and hepatocellular carcinoma, but supporting evidence is based on indirect data. Few publications have examined the results of repeat liver biopsies to evaluate progression of fibrosis.
The aims of this study were to assess rate of fibrosis progression in untreated patients with nonalcoholic steatohepatitis and to identify associated variables.
Among 106 patients, a second liver biopsy was proposed to those who had undergone their first liver biopsy at least 3 years before. None of them had been given pharmacological therapy. Liver biopsy samples were evaluated blindly. Variables were compared between patients with (group P) and without (group NP) fibrosis progression, using a Wilcoxon rank-sum test for numerical variables and a difference of two binomial proportions for categorical ones.
Twenty-two patients (median age, 45 years; age range, 20-69 years; 13 women; diabetes in 8 patients, obesity in 10 patients) underwent a second liver biopsy 4.3 years (range, 3.0-14.3 years) after the first.
Fibrosis progression was found in 7 patients in group P (31.8%), no progression was found in 15 patients in group NP. There were no differences between both groups regarding age, gender, diabetes, hyperlipidemia, ALT levels, AST-to-ALT ratio levels, albumin levels, prothrombin activity, steatosis, or inflammation.
Obesity was significantly more prevalent in group P (86%) than in group NP (27%; P = .01). Basal body mass index was higher in group P (median, 33.2; range, 29.1-38.2) than in group NP (median, 29.0; range, 24.0-38.1; P = .024). Time between biopsies was not different between groups.
In conclusion, progression of liver fibrosis was found in a third of nonalcoholic steatohepatitis patients 4.3 years after the first liver biopsy, and obesity and body mass index were the only associated factors with such progression.
Article Text
Nonalcoholic fatty liver disease is probably the main cause of chronic liver disease in the West. The histological spectrum of nonalcoholic fatty liver disease includes simple steatosis (type 1), steatosis plus lobular inflammation (type 2), steatosis plus ballooning degeneration (type 3), and steatosis plus ballooning degeneration plus Mallory bodies or fibrosis (type 4). This classification is clinically important. Whereas simple steatosis seems to be a benign and nonprogressive condition, nonalcoholic steatohepatitis (NASH) that includes the types 3 and 4 of nonalcoholic fatty liver disease is recognized as a potentially progressive disease that may cause cirrhosis and liver-related death.
However, our knowledge of the natural history of NASH is still very limited and is largely based on indirect evidence. Cross-sectional series have shown that 30% to 40% of patients have advanced liver fibrosis at the time of presentation, whereas 10% to 15% of them may have established cirrhosis. Three studies have found that patients with cryptogenic cirrhosis have a greater rate of diabetes, obesity, or both than those with cirrhosis resulting from other causes, suggesting that cryptogenic cirrhosis may represent burned out NASH. In patients who underwent liver transplantation for cryptogenic cirrhosis, steatosis, and NASH have been shown in the graft during follow-up. Finally, the appearance of hepatocellular carcinoma has been reported in NASH patients, and in two large series of patients with hepatocellular carcinoma evaluated retrospectively, it has been found that cryptogenic cirrhosis (with the clinical phenotype of NASH) was the underlying liver disease in 7% to 13% of the cohort.
Therefore, indirect data suggest that NASH may cause the entire spectrum of complications of chronic liver disease: progressive fibrosis, cirrhosis, end-stage liver disease, and hepatocellular carcinoma. However, longitudinal studies showing what percentage of patients with NASH will have a progressive course are lacking. To our knowledge, the published results of repeat liver biopsies come from only 56 patients without cirrhosis but with NASH (included in six different studies). The second biopsies were performed 1.2 to 15.7 years after the first and showed fibrosis progression in 39.3% of patients. The first four studies were clinical series examining NASH in which only a minority of patients underwent a repeat biopsy, whereas the last two studies were specially designed to evaluate histological changes. Furthermore, none of the previous studies addressed whether basal variables could differentiate patients with and without progression. These previous results are very interesting, but the number of patients studied is still very limited, and further prospective studies should be conducted to expand the knowledge of the natural history of NASH.
Thus, the aims of this prospective, longitudinal study were to assess the progression rate of liver fibrosis in nontreated NASH patients and to identify clinical, biochemical, and histological variables associated with fibrosis progression.
AUTHOR DISCUSSION
Our study shows that 31.8% of 22 patients with NASH had a progression of liver fibrosis over a median follow-up of 4.3 years. These findings confirmed the progressive potential of NASH. However, most of the patients (68.2%) had no fibrosis progression. The different evolution can not be explained by a smaller time span between biopsies in the NP group. These patients underwent their second liver biopsy at a median time of 4.3 years after the first (range, 3.2-7.9 years), not significantly different from those in the P group.
It is important to emphasize that both the pathogenesis and the natural history of nonalcoholic fatty liver disease and NASH are still incompletely understood, but they may be closely related. It has been postulated that most of the patients with nonalcoholic fatty liver disease have a fatty liver alone, a smaller proportion has steatohepatitis, and a percentage has advanced fibrosis. A two-hit hypothesis has been proposed to explain the mechanisms of liver damage in NASH. The first hit has been associated with insulin resistance, which has been demonstrated in almost all NASH patients studied and would cause the accumulation of excess fat in the hepatocytes. However, in only a proportion of patients with insulin resistance and NASH will an advanced liver disease develop, and other factors, or hits (like oxidative stress, mitochondrial dysfunction, and abnormal cytokine production), should be present to produce severe cellular injury or fibrosis.
Considering that only approximately one third of NASH patients will have increasing liver fibrosis in the midterm, it would be useful to know the factors associated with that evolution, and our study sought to identify clinical, biochemical, and histological variables able to predict progression. Among 12 variables studied at baseline, we found that the prevalence of obesity and BMI were the only variables that were significantly different between the groups (with 86% and 27% of patients being obese in the groups P and NP, respectively). Interestingly, the changes in BMI during the follow-up were not different between the groups. In fact, the gradients (final vs. basal values) of BMI were very close to 0 in both groups. Thus, among NASH patients, those with obesity would be very prone to fibrosis progression. Furthermore, in this subgroup of patients, the dietary recommendations seem to be insufficient to prevent progression, and they should be included in trials of experimental pharmacological therapy (or immediately treated when a drug is accepted as being efficacious). Presence of diabetes has been shown to be an independent predictor of severe fibrosis in a retrospective study that analyzed a large number of patients. Among our patients, four of seven (57%) from group P and 4 of 15 (27%) from group NP had diabetes. The P value (.2267) was far from being significant, but we can not exclude a type II error because of the small sample size of patients having both NASH and diabetes.
All the patients were referred to the Nutrition Department and encouraged to follow a low-calorie and low-fat diet. However, it is known that it is difficult for these patients to adhere to dietary therapy in the long term. During the follow-up, 27% of patients (6 of 22) achieved a 5% or more reduction in their body weight, and that percentage was not different between groups P and NP. Thus, we believe that it is unlikely that the dietary treatment could have played an important role in modifying the evolution of our patients' courses, but we can not exclude a mild beneficial effect. Among the four patients showing a decrease in the final fibrosis score, only one had had that kind of weight reduction.
Comparing our results with those of 59 patients without cirrhosis but with NASH and repeat liver biopsies previously published in the literature, we found in our study a lower rate of fibrosis progression (32%) during a longer follow-up (median, 4.3 years; mean, 5.3 years) than those observed in the first four studies that were clinical series and included 30 patients. It is difficult to determine the reasons for that difference, because the demographical or analytical data are not fully available in the papers. However, among 24 patients whose clinical data are known, 22 (91.7%) were obese (the only predictor of progression, according to our results), and this could be the main explanation for a faster progression. The last two studies, like ours, were designed to analyze changes in histological findings, and Harrison et al. reported results impressively similar: 7 of their 22 patients showed fibrosis progression in a mean time of 5.7 years between biopsies. The clinical data of their patients also were similar to ours, but the obesity prevalence was slightly higher (77%).
Other findings of our study worth noting are as follows. First, the inflammatory activity in the basal liver biopsy was not different between patients with or without fibrosis progression, in contrast to what was expected. However, it should be noted that the grade of inflammation was mild in most of the patients (19 of 22). Second, changes in grade of steatosis run an independent course from those in fibrosis, being found in the final liver biopsy a decrease in steatosis and an increase in fibrosis, either in the entire population or in group P patients. This observation is consistent with the suggestion that NASH patients may lose the fatty infiltration when they reach the stage of cirrhosis, which becomes a cryptogenic cirrhosis. Third, the evolution of ALT values was not different between both groups and, especially in group P, a normalization of ALT values was observed in six of seven patients. It is important to emphasize that a normalization of ALT levels does not guarantee fibrosis stabilization or improvement, because many pilot studies evaluating pharmacological treatments in NASH have claimed different drugs to be efficacious based only on the decreasing ALT values.
The rate of progression of liver fibrosis has not been studied previously in NASH. Considering the entire population, the rate of liver fibrosis progression was estimated in 0.059 units of fibrosis per year. In a recent study, in patients with chronic hepatitis C, Ghany et al. found progression of liver fibrosis in 39% over a mean interval of 44 months (range, 2-211 mo) between both biopsies. The rate of liver fibrosis progression was estimated in 0.12 fibrosis units per year. Thus, our figure of fibrosis progression in NASH patients is approximately half of that found in hepatitis C patients. However, in patients from group P, the rate of progression was estimated to be 0.280 units of fibrosis per year.
Some very recent, noncontrolled studies have shown that insulin-sensitizing agents (rosiglitazone and pioglitazone) can lead to improvement in histological features in NASH patients after 48 weeks of treatment. If these results were confirmed in larger, controlled studies, this kind of therapy would be accepted as being able to modify the natural evolution of NASH and would be recommended promptly for clinical practice. In that eventuality, studies like ours, aimed at understanding natural history, no longer would be carried out because they would be considered unethical, and the knowledge of the natural history of NASH would remain rather limited.
In summary, the results of this longitudinal study have shown progression of liver fibrosis in approximately one third of patients with NASH in a median follow-up of 4.3 years, the presence of obesity being the only factor associated with the progression. Patients with NASH who are obese should be included in controlled trials of experimental pharmacological therapy or should be treated promptly when some agreement exists regarding the efficacy of any drug in retarding the progression of fibrosis.
Patients and Methods
From October 1986 through December 2002, 106 patients were diagnosed with NASH in the Liver Unit at the Professor Alejandro Posadas Hospital. The diagnosis of NASH was based on the following four criteria. Persistently abnormal alanine aminotransferase (ALT) levels, aspartate aminotransferase (AST) levels, or both was the first criterion. A daily alcohol intake of less than 40 g in men and less than 20 g in women, as confirmed by patient anamnesis and interview of close family members, was the second criterion. These cutoff levels of ethanol intake were chosen because they were the usual ones when we designed the study. Surrogate biochemical markers of alcohol consumption were not used. Appropriate exclusion of other causes of chronic liver disease, such as hepatitis B and C, autoimmune hepatitis, drug-induced hepatitis, primary biliary cirrhosis, hemochromatosis, Wilson disease, was the third criterion. The fourth criterion was characteristic features in the liver biopsy, including macrovesicular steatosis (>10% of hepatocytes) and lobular inflammation plus ballooning degeneration, Mallory hyaline fibrosis, sinusoidal fibrosis, or a combination thereof. Patients demonstrating only steatosis and lobular inflammation in their biopsy results were not considered to have NASH. Either ballooning degeneration or sinusoidal fibrosis had to be present to confirm the diagnosis of NASH. This population included 55 males and 51 females with a median age of 45 years (range, 17-78 years).
Patients were considered to have diabetes mellitus either if they were receiving insulin or oral hypoglycemic treatment or if a fasting plasma glucose test result was 126 mg/dL or more, according to the American Diabetes Association definition. Hyperlipidemia was diagnosed when fasting levels of cholesterol were more than 200 mg/dL, fasting levels of triglycerides were more than 200 mg/dL in two occasions, or both. Obesity was defined as a body mass index (BMI) of more than 30, both in men and women. BMI was calculated using the following formula: kg(weight)/m2(height). Waist circumference or waist-to-hip ratio measurements were not obtained in these patients.
All the patients were referred to the Nutrition Department for the treatment of their metabolic disorders (hyperlipidemia, obesity, glucose intolerance, or diabetes), but no experimental pharmacological treatment for NASH (e.g. ursodeoxycholic acid, vitamin E, glitazones) was given to any patient. No patient was taking drugs associated with secondary NASH, such as corticosteroids, perhexiline, tamoxifen, and amiodarone.
All the patients were scheduled to undergo a repeat liver biopsy at least 3 years after the previous biopsy. Patients lost to follow-up were contacted by phone calls or conventional mailing. The study protocol conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the ethics and research committees of our hospital. Informed consent was obtained from each enrolled patient.
Liver biopsy samples were obtained by the percutaneous route using the Menghini method. One or two passes were performed to assure samples at least 25 mm in length. Formalin-fixed, paraffin-embedded liver sections were stained routinely with hematoxylin and eosin, silver reticulin, Masson trichrome, Perls' Prussian blue, and diastase-resistant periodic acid-Schiff.
After determining patient inclusion, all the liver biopsy specimens (basal and final samples) were reexamined in a blind and nonpaired manner by an experienced pathologist (E.A.) who was unaware of the clinical and biochemical data of the patients or the order of the biopsies. Studies on the liver specimens included a semiquantitative assessment of the grades of steatosis (mild or grade 1, >10% but <33% of hepatocytes affected; moderate or grade 2, 33%-66% of hepatocytes affected; severe or grade 3, >66% of hepatocytes affected); of inflammatory activity (according to Brunt classification); and of fibrosis (according to Brunt classification and to Ishak classification). Brunt classification of fibrosis assessment includes five stages: stage 0, no fibrosis; stage 1, zone 3 perisinusoidal or pericellular fibrosis, focally or extensively present; stage 2, zone 3 perisinusoidal or pericellular fibrosis with focal or extensive periportal fibrosis; stage 3, zone 3 perisinusoidal or pericellular fibrosis and portal fibrosis with focal or extensive bridging fibrosis; stage 4, cirrhosis. Ishak classification ranges from zero to six stages: stage 0, no fibrosis; stage 1, fibrous expansion of some portal areas, with or without short fibrous septa; stage 2, fibrous expansion of most portal areas, with or without short fibrous septa; stage 3, fibrous expansion of most portal areas with occasional portal to portal bridging; stage 4, fibrous expansion of portal areas with marked bridging (portal to portal) as well as portal to central; stage 5, marked bridging with occasional nodules (incomplete cirrhosis); stage 6, cirrhosis, probable or definite.
Progression of liver fibrosis was defined as an increase 1 grade or more in the final stage with respect to the basal biopsy, in any of the two classifications. Although Brunt classification was designed especially for NASH patients, we decided to analyze changes in fibrosis according to Ishak classification as well, because of its greater flexibility in evaluating septal fibrosis (score 3 for occasional bridging, score 4 for marked bridging), incomplete cirrhosis (score 5), and definite cirrhosis (score 6).
|
|
| |
| |
|
|
|
