Abstract
Background
Although there is unequivocal evidence for progression of nonalcoholic steatohepatitis (NASH) to cirrhosis, there is uncertainty with regard to the progression to nonalcoholic fatty liver (NAFL) and NASH.
Aims
We investigated the rate of progression to NASH and advanced fibrosis in patients with nonalcoholic fatty liver disease (NAFLD) and assessed the factors associated with such progression.
Methods
Histological assessment was performed in 36 patients with NAFLD with paired liver biopsies (≥1 year apart; median, 3.8 years; range, 1–10.33 years). NASH Clinical Research Network (NASH CRN) criteria were used to assess NAFLD Activity Score (NAS).
Results
At baseline, 26 (72%) patients had NAFL and 10 (28%) patients had NASH. At follow-up, 27% NAFL progressed to NASH (NAS score ≥5), and 50% of patients with NASH no longer met the criteria of NASH. Fibrosis progressed in 15 (42%), regressed in 9 (25%), and remained stable in 12 (33%) patients overall. Thirty-five percent of patients with NAFL had fibrosis progression. The incidence of type 2 diabetes mellitus (T2DM) was higher in patients with NASH versus NAFL (40% vs. 27%). Both at the time of baseline and follow-up, liver biopsies, composite models of noninvasive scores such as Fibrosis-4 (FIB-4) score and NAFLD fibrosis score, and ratio of aspartate aminotransferase (AST) to alanine aminotransferase (ALT) were all significantly higher in progressors than in nonprogressors.
Conclusions
NAFLD is a dynamic liver disease with varying degrees of progression and regression. T2DM was strongly associated with fibrosis progression. Noninvasive fibrosis scores such as AST/ALT ratio, FIB-4 score, and NAFLD fibrosis score can identify those at risk of fibrosis progression.
Keywords: nonalcoholic fatty liver disease, nonalcoholic steatohepatitis, nonalcoholic fatty liver, obesity-related liver disease, metabolic syndrome, natural history
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; FIB-4 score, Fibrosis-4 score; NAFL, Nonalcoholic fatty liver; NAFLD, Nonalcoholic fatty liver disease; NAS, NAFLD Activity Score; NASH, Nonalcoholic steatohepatitis; NASH CRN, NASH Clinical Research Network; T2DM, Type 2 diabetes mellitus
Nonalcoholic fatty liver disease (NAFLD) is one of the most common liver diseases in the world, the prevalence of which is steadily rising.1 As per studies, 25.24% people globally are thought to be affected by NAFLD.1 With the obesity epidemic, the incidence of metabolic syndrome and NAFLD is also on the rise.2 As previous studies have hypothesized, insulin resistance is the driving force behind NAFLD/nonalcoholic steatohepatitis (NASH).2
NAFLD is defined as the presence of increased hepatic steatosis in the absence of other causes of secondary hepatic fat deposition such as alcohol abuse, endocrine disorders, chronic hepatitis C virus (HCV) infection, medications, or genetic disorders.3 NAFLD is a dynamic spectrum of liver disease, with nonalcoholic fatty liver (NAFL) on one end of the spectrum through NASH with and without advanced fibrosis and cirrhosis on the other end. NAFL is described as the presence of ≥5% hepatic steatosis with no evidence of hepatocellular injury.3 NASH is described as the presence of ≥5% hepatic steatosis and lobular inflammation with hepatocyte injury (hepatocyte ballooning), with or without any fibrosis.3
Along the spectrum, liver fibrosis is the histologic feature that predicts morbidity.4 The degree of fibrosis in patients with NAFLD appears to predict not only liver-related mortality but also cardiovascular and all-cause mortality.5 NAFLD/NASH is fast becoming the most common cause of hepatocellular carcinoma (HCC).6 A recent study showed that over a time period of 3.2 and 7.2 years, 2.4% and 12.8% patients with NASH cirrhosis develop HCC, respectively.6 NASH is the second most common indicator for liver transplant overall and is the leading cause for liver transplant in females in the United States.7
Despite the high prevalence of NAFLD, only a few studies have assessed the histological progression of NAFL, NASH, and fibrosis. The aim of our study was to determine (1) the rate of progression of NAFL to NASH and (2) the rate of progression of fibrosis in the entire cohort of patients with fatty liver with the help of paired biopsies.
Methods
Patient population
This study cohort was identified at Methodist University Hospital in Memphis, Tennessee, a large tertiary care hospital at Memphis metropolitan area. Figure 1 summarizes the cohort of patients included in the study. A retrospective review of Electronic Medical Records (EMR) identified 5696 consecutive adult patients (≥18 years) with at least one liver biopsy between January 2006 and December 2016. After exclusion of patients with significant alcohol abuse (n = 641), HIV/AIDS (n = 54), hepatitis B (n = 215), hepatitis C (n = 1026), 3749 remaining patients were reviewed for diagnosis of NAFLD using International Classification of Diseases diagnosis codes-9 and 10 (ICD-9/ICD-10). We identified 1271 unique NAFLD patients using this search criteria. Pathology reports were reviewed manually to confirm NAFLD in these patients to form the final cohort of 942 patients with NAFLD. For the patients with ≥2 liver biopsies, the first and the most recent biopsies were used. If the patient took part in a therapeutic clinical trial, the biopsy before the beginning of the clinical trial was used as the second biopsy. None of the patients received therapeutic clinical trial medications during the study duration. From the final cohort, 36 patients with ≥2 liver biopsies were included after exclusion of 3 patients who had biopsies within one year (median time interval, 45.2 months; range, 12–124 months).
Figure 1.
Patients in cohort diagram. NAFLD, nonalcoholic fatty liver disese.
The laboratory values were gathered retrospectively from the time of liver biopsy or within 1 year. We defined liver biopsy specimen as adequate if the length of the reported specimen is ≥10 mm and/or the number of portal tracts was ≥7. The clinical details that were obtained include gender, age, weight, height, ethnicity, aspartate aminotransferase (AST), ALT, platelets, total bilirubin, international normalized ratio (INR), alkaline phosphatase, and albumin. Patients were identified as having type 2 diabetics (T2DM) if they had HbA1c > 6.5 and received anti-diabetic medications and/or insulin treatment. The aforementioned laboratory values were used to calculate the FIB-4, AST to Platelet Ratio Index (APRI), AST/ALT ratio, and NAFLD fibrosis scores.8
Histological assessment
Data from liver biopsies were manually extracted by chart review by S.G., Y.K.R. and H.K.M. All biopsies were a minimum of 10 mm in length and considered adequate based on our predefined criteria. The number of portal tracts was not consistently reported and varied from 3–13. The liver pathologists reported the results using the NASH Clinical Research Network (NASH CRN) criteria.9 The NAFLD Activity Score (NAS) was recorded from 0 to 8. The NAS score grades steatosis (0–3), lobular inflammation (0–3), and hepatocellular ballooning (0–2). NAFL was defined as having only steatosis and/or steatosis with mild lobular inflammation without hepatocyte ballooning. If the score was ≥5, the patient was classified as having NASH.10 Fibrosis was staged from 0 to 4.
Statistical analysis
Continuous distributed variables were characterized as mean ± standard deviation (SD). Categorical variables were characterized as median and range. Student's t test was used to compare means of normally distributed continuous variables. Chi-squared or Fisher exact test was used to determine the distribution of categorical variables as appropriate. To compare the nonnormally distributed continuous variables, medians were compared using the Mann-Whitney U test. Statistical analysis was performed using SAS 9.3 (SAS Institute, Cary NC).
Results
As previously discussed, from a total of 942 patients with histologically confirmed NAFLD, 36 patients with at least 2 liver biopsies were identified more than 1 year apart. This cohort of 36 patients was used for subsequent analysis.
Clinical and biochemical characteristics at baseline and follow-up are summarized in Table 1. A majority of the patients in the cohort were female (n = 30). The mean age at baseline was 48.4 ± 10.92 years, and the mean body mass index (BMI) was 33.9 ± 6.15 kg/m2. At baseline, 26 (72%) patients had NAFL and 10 (28%) patients had NASH. Of the 26 patients with NAFL, 7 patients (27%) progressed to NASH and 5 patients (50%) with NASH no longer met the criteria of NASH (NAS score ≥ 5) at follow-up. Overall fibrosis progressed in 15 (42%) patients; 9 (25%) patients progressed by 1 stage, 2 (6%) patients progressed by 2 stages, 3 (8%) patients progressed by 3 stages, and 1 (3%) patient progressed by 4 stages. Overall regression of fibrosis occurred in 9 (25%) patients; 1 (3%) patient regressed by 1 stage, 6 (17%) patients regressed by 2 stages, and 2 (6%) patients regressed by 3 stages. There was no change in fibrosis in 12 (33%) patients. Of the patients with NAFL, 35% patients had fibrosis progression. After a median follow-up of 3.8 years, the prevalence of diabetes increased to 55.5% from 30.5% (P < 0.002) and the serum creatinine increased significantly from a baseline of 1.30 ± 2.20 to 1.48 ± 2.67 (P = 0.012).
Table 1.
Comparison of Demographics and Clinical Factors at Baseline and at Follow-up Liver Biopsy.
| Characteristic | Baseline (N = 36) | Follow-up (N = 36) | P value |
|---|---|---|---|
| Age (years) | 48.4 ± 10.92 | 52.83 ± 11.17 | <0.001a |
| Gender (% male) | 16.67% | 16.67% | 1.00b |
| BMI (kg/m2) | 33.9 ± 6.15 | 34.13 ± 6.15 | 0.74a |
| T2DM | 30.5% | 55.56% | <0.002b |
| Bilirubin (mg/dl) | 0.84 ± 0.90 | 0.65 ± 0.41 | 0.23a |
| Alkaline phosphatase (IU/l) | 88.59 ± 33.19 | 101.46 ± 53.92 | 0.25a |
| Albumin (g/dl) | 3.92 ± 0.70 | 4.02 ± 0.59 | 0.26a |
| Creatinine (mg/dl) | 1.30 ± 2.20 | 1.48 ± 2.67 | 0.012a |
| INR | 1.07 ± 0.16 | 1.03 ± 0.07 | 0.69a |
| ALT (IU/l) | 71.16 ± 64.3 | 59.96 ± 58.51 | 0.93a |
| AST (IU/l) | 54.5 ± 36.11 | 40.21 ± 24.51 | 0.17a |
| Platelets (×109 per liter) | 246.62 ± 76.22 | 223.66 ± 70.84 | 0.12a |
| AST/ALT ratio | 0.96 ± 0.59 | 0.90 ± 0.54 | 0.47a |
| FIB-4 score | 1.52 ± 0.88 | 1.66 ± 1.23 | 0.36a |
| NAFLD fibrosis score | −1.08 ± 1.56 | −0.51 ± 1.65 | 0.09a |
| APRI score | 0.67 ± 0.60 | 0.56 ± 0.49 | 0.71a |
| NAS score | |||
| Mean ± SD | 3.61 ± 1.76 | 3.22 ± 1.94 | 0.36c |
| Median (IQR) | 4 (3–5) | 3 (2–5) | |
| Steatosis | 2 (1–3) | 1 (1–2) | |
| Lobular | 1 (1–2) | 1 (0–1) | |
| Inflammation ballooning | 0 (0–2) | 1 (0–2) | |
| Fibrosis stage | |||
| Mean ± SD | 1.33 ± 1.17 | 1.53 ± 1.66 | 0.66c |
| Median, (IQR) | 1.5 (0–2) | 1 (0–3) | |
| 0 | 13 | 17 | |
| 1 | 5 | 3 | |
| 2 | 11 | 3 | |
| 3 | 7 | 6 | |
| 4 | 0 | 7 | |
BMI, body mass index; T2DM, type II diabetes mellitus; ALT, alanine aminotransferase; AST, aspartate aminotransferase; INR, international normalized ratio; NAS, NAFLD Activity Score; FIB-4 score, Fibrosis-4 score; APRI score, AST to Platelet Ratio Index; AST/ALT ratio, aspartate aminotransferase/alanine aminotransferase ratio; NAFLD score, Nonalcoholic Fatty Liver Disease Fibrosis score.
Student's t test.
Chi-squared.
Wilcoxon rank test.
In Table 2, patients with NAFL and patients with NASH were compared at baseline. The average age for both is approximately 48 years. In the current cohort, the entire NASH population consisted of females. Characteristics such as BMI and bilirubin, albumin, INR, ALT, AST, and platelet levels were not different in both subsets of patients. However, patients with NASH had a baseline creatinine 0.74 ± 0.13 and NAFL had a baseline creatinine 1.52 ± 2.56 (P = 0.036).
Table 2.
Comparisons of Patients With NASH With Those of NAFL at Baseline.
| Characteristic | NAFL (N = 26) | NASH (N = 10) | P value |
|---|---|---|---|
| Age (years) | 48.19 ± 11.30 | 49 ± 10.4 | 0.91a |
| Gender (% male) | 23% | 0% | 0.16b |
| BMI (kg/m2) | 33.98 ± 6.48 | 34.02 ± 5.52 | 0.92a |
| T2DM | 26.92% | 40% | 0.45b |
| Bilirubin (mg/dl) | 0.80 ± 0.53 | 0.94 ± 1.51 | 0.12a |
| Alkaline phosphatase (IU/l) | 87.43 ± 34.83 | 91.56 ± 30.29 | 0.68a |
| Albumin (g/dl) | 3.89 ± 0.79 | 4.01 ± 0.35 | 0.89a |
| Creatinine (mg/dl) | 1.52 ± 2.56 | 0.74 ± 0.13 | 0.036a |
| INR | 1.03 ± 0.11 | 1.15 ± 0.23 | 0.07a |
| ALT (IU/L) | 71.17 ± 70.80 | 71.11 ± 47.31 | 0.68a |
| AST (IU/L) | 52.83 ± 39.38 | 58.59 ± 27.52 | 0.17a |
| Platelets (×109 per liter) | 247.08 ± 85.24 | 245.33 ± 46.54 | 0.86a |
| AST/ALT ratio | 0.88 ± 0.32 | 1.18 ± 1.0 | 0.59a |
| FIB-4 score | 1.45 ± 0.89 | 1.72 ± 0.86 | 0.32a |
| NAFLD fibrosis score | −1.28 ± 1.55 | 0.53 ± 1.55 | 0.28a |
| APRI score | 0.68 ± 0.68 | 0.63 ± 0.31 | 0.32a |
BMI, body mass index; T2DM, type II diabetes mellitus; ALT, alanine aminotransferase; AST, aspartate aminotransferase; INR, international normalized ratio; NAS, NAFL Activity Score; FIB-4 score, Fibrosis-4 score; APRI score, AST to Platelet Ratio Index; AST/ALT ratio, aspartate aminotransferase/alanine aminotransferase ratio; NAFLD score, Nonalcoholic Fatty Liver Disease Fibrosis Score.
Wilcoxon rank test.
Fisher's Exact test.
Histologically, 26 patients with NAFL had a mean NAS score of 3 (1–8) and 10 patients with NASH had a mean NAS score of 5.5 (1–8) (Table 3). The median time between the initial and follow-up biopsy was 3.8 years (range, 1–10.33 years). There was a 50% fibrosis progression in patients with NASH. At baseline, 80% of patients with NASH had a fibrosis score of ≤2 and 20% patients with NASH had fibrosis score of 3 and 4. On follow-up, 58% of patients with NASH had a fibrosis score of ≤2 and 42% of patients with NASH had a fibrosis score of 3 and 4. Figure 2 illustrates the distribution of fibrosis stages at baseline vs follow-up in patients with NAFL and patients with NASH.
Table 3.
Comparisons of Patients With NASH and With Those of NAFL at Baseline and Follow-up.
| Characteristic | NAFL (N = 26) | NASH (N = 10) | P value |
|---|---|---|---|
| Baseline NAS | |||
| (Mean ± SD) | 2.88 ± 1.37 | 5.6 ± 0.70 | <0.001a |
| (Median, IQR) | 3 (2–4) | 5 (5–6) | |
| Steatosis | |||
| (Mean ± SD) | 1.58 ± 0.95 | 2.8 ± 0.42 | 0.002a |
| (Median, IQR) | 2 (1–2) | 3 (3–3) | |
| Lobular inflammation | |||
| (Mean ± SD) | 0.85 ± 0.67 | 1.78 ± 0.44 | 0.002a |
| (Median, IQR) | 1 (0–1) | 2 (2–2) | |
| Ballooning | |||
| (Mean ± SD) | 0.42 ± 0.58 | 1 ± 0.87 | 0.07a |
| (Median, IQR) | 0 (0–1) | 1 (0–2) | |
| Baseline fibrosis | |||
| (Mean ± SD) | 1.08 ± 1.20 | 2.00 ± 0.82 | 0.048a |
| (Median, IQR) | 1 (0–2) | 2 (2–2) | |
| 0 | 12 (50%) | 1 (10%) | |
| 1 | 5 (19%) | 0 | |
| 2 | 4 (15%) | 7 (70%) | |
| 3 | 5 (19%) | 2 (20%) | |
| 4 | 0 | 0 | |
| Follow-up | NAFL (24) | NASH (12) | P value |
| Follow-up NAS | |||
| (Mean ± SD) | 2.1 ± 1.3 | 5.4 ± 0.5 | <0.001a |
| (Median, IQR) | 2 (1–3) | 5 (5–6) | |
| Steatosis | |||
| (Mean ± SD) | 1.0 ± 0.69 | 2.4 ± 0.79 | 0.004a |
| (Median, IQR) | 1 (1–1) | 3 (2–3) | |
| Lobular inflammation | |||
| (Mean ± SD) | 0.54 ± 0.51 | 1.42 ± 0.51 | 0.008a |
| (Median, IQR) | 1 (0–1) | 1 (1–2) | |
| Ballooning | |||
| (Mean ± SD) | 0.54 ± 0.72 | 1.58 ± 0.51 | 0.001a |
| (Median, IQR) | 0 (0–1) | 2 (1–2) | |
| Follow-up fibrosis | |||
| Mean ± SD | 1.4 ± 1.8 | 1.8 ± 1.4 | 0.37a |
| (Median, IQR) | 0 (0–3.5) | 2.0 (0.5–3) | |
| 0 | 14 (58.3%) | 3 (25%) | |
| 1 | 1 (4.1%) | 2 (16.6%) | |
| 2 | 1 (4.1%) | 2 (16.6%) | |
| 3 | 2 (7.6%) | 4 (33.3%) | |
| 4 | 6 (23%) | 1 (8.3%) | |
| Fibrosis progression | 34.62% | 50% | 0.46a |
| Fibrosis rate (stage/year) | 0.03 ± 0.73 | −0.2 ± 0.81 | 0.54a |
| Time between biopsies (year) | |||
| Mean ± SD | 4.6 ± 2.96 | 4.0 ± 2.49 | |
| Median (range) | 4 (1–10) | 4 (1–8) | 0.66a |
Wilcoxon rank test.
Figure 2.
Constituent stages of fibrosis in NAFL and NASH at baseline and follow-up, 0 = fibrosis stage 0; 1 = fibrosis stage 1; 2 = fibrosis stage 2; 3 = fibrosis stage 3; 4 = fibrosis stage 4. NASH, nonalcoholic steatohepatitis; NAFL, nonalcoholic fatty liver.
Patients with progression of fibrosis were compared with patients whose fibrosis remained stable or regressed to determine factors that were linked with fibrosis progression. As shown in Table 4, the composite models such a FIB-4 score (1.95 ± 0.96 vs 1.24 ± 0.73 P = 0.02), NAFLD fibrosis score (−0.29 ± 1.24 vs −1.61 ± 1.56 P = 0.019), and AST/ALT ratios (1.05 ± 0.25 vs 0.91 ± 0.72, P = 0.018) were all significantly higher in progressors than in nonprogressors at baseline liver biopsies. After a median follow-up of 3.8 years, FIB-4 score (2.36 ± 1.49 vs 1.27 ± 0.89, P = 0.02), NAFLD fibrosis score (0.72 ± 0.96 vs −1.19 ± 1.56, P = 0.005), and AST/ALT ratios (1.33 ± 0.68 vs 0.67 ± 0.24, P = 0.005) continued to remain significantly higher. A significantly higher number of progressors had diabetes (30.5% vs 25%, P = 0.04) at follow-up.
Table 4.
Clinical Factors at Baseline and Follow-up Biopsy for All Patients and a Comparison Between Patient With Fibrosis Progression and Without Fibrosis Progression.
| Characteristic | All patients (N = 36) | No progression of fibrosis (N = 21) | Progression of fibrosis (N = 15) | P value |
|---|---|---|---|---|
| Results at baseline biopsy | ||||
| Age (years) | 48.4 ± 10.92 | 45.77 ± 9.88 | 52.57 ± 11.53 | 0.12a |
| Gender (% male) | 16.67% | 13.89% | 2.78% | 0.37b |
| BMI (kg/m2) | 33.9 ± 6.15 | 34.41 ± 6.67 | 33.33 ± 5.42 | 0.80a |
| T2DM | 30.5% | 13.89% | 16.67% | 0.27b |
| Bilirubin (mg/dl) | 0.84 ± 0.90 | 0.99 ± 1.11 | 0.60 ± 0.23 | 0.69a |
| Alkaline phosphatase (IU/l) | 88.59 ± 33.19 | 85.40 ± 32.54 | 93.92 ± 35.01 | 0.42a |
| Albumin (g/dl) | 3.92 ± 0.70 | 3.93 ± 0.76 | 3.91 ± 0.62 | 0.98a |
| Creatinine (mg/dl) | 1.30 ± 2.20 | 0.95 ± 0.27 | 1.80 ± 3.42 | 0.24a |
| INR | 1.07 ± 0.16 | 1.07 ± 0.2 | 1.07 ± 1.11 | 0.20a |
| ALT (IU/l) | 71.16 ± 64.3 | 82.25 ± 73.59 | 52.67 ± 41.26 | 0.19a |
| AST (IU/l) | 54.5 ± 36.11 | 56.85 ± 38.93 | 50.67 ± 32.09 | 0.48a |
| Platelets (×109 per liter) | 246.62 ± 76.22 | 259.05 ± 70.96 | 228.96 ± 82.53 | 0.23a |
| AST/ALT ratio | 0.96 ± 0.59 | 0.91 ± 0.72 | 1.05 ± 0.25 | 0.018a |
| FIB-4 score | 1.52 ± 0.88 | 1.24 ± 0.73 | 1.95 ± 0.96 | 0.02a |
| NAFLD fibrosis score | −1.08 ± 1.56 | −1.61 ± 1.56 | −0.29 ± 1.24 | 0.019a |
| APRI score | 0.67 ± 0.60 | 0.67 ± 0.69 | 0.66 ± 0.46 | 0.64a |
| NAS score | ||||
| (Mean ± SD) | 3.61 ± 1.76 | 3.73 ± 1.61 | 3.43 ± 2.03 | |
| (Median, IQR) | 4 (3–5) | 4 (3–4) | 4 (2–5) | 0.88a |
| Steatosis | 2 | 2 | 2 | |
| Lobular inflammation | 1 | 1 | 1 | |
| Ballooning | 0 | 0.5 | 0 | |
| Fibrosis stage | ||||
| (Mean ± SD) | 1.33 ± 1.17 | 1.41 ± 1.22 | 1.21 ± 1.12 | |
| (Median, IQR) | 1.5 (0–2) | 2 (0–2) | 1 (0–2) | 0.65a |
| 0 | 13 (36%) | 8 (36.3%) | 5 (35.7%) | |
| 1 | 5 (13.8%) | 2 (9%) | 3 (21.4%) | |
| 2 | 11 (30.5%) | 7 (31.8%) | 4 (28.5%) | |
| 3 | 7 (19.4%) | 5 (22.7%) | 2 (14.2%) | |
| 4 | 0 | 0 | 0 | |
| Results at follow-up biopsy | ||||
| BMI (KG/M2) | 34.13 ± 6.15 | 34.13 ± 6.03 | 34.13 ± 6.57 | 0.87a |
| T2DM | 55.6% | 25% | 30.5% | 0.04b |
| Bilirubin (mg/dl) | 0.65 ± 0.41 | 0.69 ± 0.47 | 0.57 ± 0.26 | 0.62a |
| Alkaline phosphatase (IU/l) | 101.46 ± 53.92 | 93.06 ± 49.82 | 116.60 ± 60.33 | 0.26a |
| Albumin (g/dl) | 4.02 ± 0.59 | 4.10 ± 0.57 | 3.87 ± 0.61 | 0.75a |
| Creatinine (mg/dl) | 1.48 ± 2.67 | 0.96 ± 0.20 | 2.47 ± 4.52 | 0.40a |
| INR | 1.03 ± 0.07 | 1.01 ± 0.05 | 1.07 ± 0.08 | 0.014a |
| ALT (IU/l) | 59.96 ± 58.51 | 71.67 ± 68.56 | 38.90 ± 17.96 | 0.20a |
| AST (IU/l) | 40.21 ± 24.51 | 40.67 ± 27.98 | 39.40 ± 17.96 | 0.87a |
| Platelets (×109 per liter) | 223.66 ± 70.84 | 237.76 ± 78.60 | 202.50 ± 53.08 | 0.13a |
| AST/ALT ratio | 0.90 ± 0.54 | 0.67 ± 0.24 | 1.33 ± 0.68 | 0.005a |
| FIB-4 score | 1.66 ± 1.23 | 1.27 ± 0.89 | 2.36 ± 1.49 | 0.02a |
| NAFLD score | −0.51 ± 1.65 | −1.19 ± 1.56 | 0.72 ± 0.96 | 0.005a |
| APRI score | 0.56 ± 0.49 | 0.52 ± 0.47 | 0.63 ± 0.54 | 0.36a |
| NAS score | ||||
| (Mean ± SD) | 3.22 ± 1.94 | 2.82 ± 1.92 | 3.86 ± 1.88 | |
| (Median, IQR) | 3 (1.5–5.0) | 3 (1–5) | 4 (3–5) | 0.13a |
| Steatosis | 1 | 1 | 1 | |
| Lobular inflammation | 1 | 1 | 1 | |
| Ballooning | 1 | 0 | 1.5 | |
| Fibrosis stage | ||||
| (Mean ± SD) | 1.53 ± 1.66 | 0.59 ± 1.22 | 3 ± 1.11 | |
| (Median, IQR) | 1 (0–3) | 0 (0–0) | 3 (2–4) | <0.001a |
| 0 | 17 | 17 (47%) | 0 (0%) | |
| 1 | 3 | 1 (2.78%) | 2 (5.56%) | |
| 2 | 3 | 1 (2.78%) | 2 (5.56%) | |
| 3 | 6 | 2 (5.56%) | 4 (11.11%) | |
| 4 | 7 | 1 (2.78%) | 6 (16.67%) | |
| Time between biopsy (yr) | 4.5 ± 3.10 | 4.29 ± 2.43 | 0.92 | |
BMI, body mass index; T2DM, type II diabetes mellitus; ALT, alanine aminotransferase; AST, aspartate aminotransferase; INR, international normalized ratio; NAS, NAFLD Activity Score; FIB-4 score, Fibrosis-4 score; APRI score, AST to Platelet Ratio Index; AST/ALT ratio, aspartate aminotransferase/alanine aminotransferase ratio; NAFLD score, Nonalcoholic fatty liver disease fibrosis score.
Wilcoxon rank test.
Fisher's Exact test.
Discussion
In this cohort of patients with NAFLD, we have demonstrated on paired liver biopsies that 27% of the patients had progression from NAFL to NASH and 50% of NASH patients no longer met the criteria for NASH. In addition, we have noted that of 15 (42%) patients who had fibrosis progression, only 3 (20%) had progressed from NAFL to NASH. A notably higher number of progressors had diabetes on follow-up. Noninvasive biomarkers such as AST/ALT ratio, FIB-4, and NAFLD fibrosis score identified patients with significant fibrosis on follow-up.
Prior studies reported that NAFL without NASH is a benign condition with little or no risk of progression.11, 12, 13 In a study by Teli et al,11 among 40 patients with simple hepatic steatosis, none had progressed to steatohepatitis or cirrhosis. It was hypothesized that in the absence of prior fibrosis or steatohepatitis, NAFL is an “extremely benign condition.” Dam-Larsen et al13 conducted a study with 109 patients with NAFL, and only one patient developed cirrhosis at follow-up which led to their hypothesis that patients with NAFL without NASH have a benign clinical course. In a study by Adams et al12 involving 103 patients with simple steatosis on initial biopsy, none progressed to fibrosis. At the time, it was concluded that patients with simple steatosis had a relatively benign course. This concept that NAFL is a benign condition has been challenged by several studies. In a prospective study by Wong et al,14 23% of patients with simple steatosis at baseline developed NASH. According to a systematic review by Pais et al15 involving 70 patients with NAFLD, 16 of the 25 patients with NAFL progressed to NASH after a mean follow-up of 3.7 years (SD ± 2.1). According to a meta-analysis involving 411 patients with NAFLD, both patients with NAFL and NASH developed liver fibrosis. The percent of patients with NAFL who had fibrosis progression was 39.1%, and 34.5% of patients with NASH had fibrosis progression. However, it was noted that patients with NAFL progressed to liver fibrosis at a slower rate. The annual fibrosis progression rate in patients with NAFL vs patients with NASH, with baseline stage 0 fibrosis, was 0.07 stages vs 0.14 stages, respectively.16 A recent study by McPherson et al17 demonstrated that 44% of patients with NAFL at baseline progressed to NASH and 37% had fibrosis progression.
We also have noted fibrosis progression in patients with NAFL without conforming to the definition of NASH by NAS criteria. In fact, in our study, 35% of patients with NAFL had fibrosis progression. Twenty-seven percent of patients with NAFL progressed to NASH. Additionally, we noted that 50% of patients with NASH had fibrosis progression, 40% of patients with NASH had fibrosis regression, and 10% patients had no change in fibrosis stage. These results were similar to those of prior studies. The study by Wong et al14 showed that among patients with NASH at baseline, 59% continued to have NASH, 35% had borderline NASH, and 6% of patients regressed to simple steatosis at 36 months. Wong et al14 demonstrated that 27% of patients overall had fibrosis progression, 48% remained stable, and 25% had fibrosis regression. The study by McPherson et al17 showed that of the patients with NASH on baseline biopsy, 93% still had NASH at follow-up, and 7% had regressed to NAFL. Forty-two percent of patients with NAFLD had progression, 40% had static disease, and 18% had regression. These findings highlight the importance of follow-up with liver biopsies in patients with NAFL, particularly with risk factors for progression of fibrosis such as diabetes. In our cohort, the smaller percentage of patients with NAFL progressing to NASH could be contributed to the lower mean NAS score (2.85 ± 1.41) at the time of initial biopsy. Another factor to take into consideration is that the patients in this cohort underwent a second liver biopsy at a median follow-up of 3.8 years compared with 6.6 years in the study by McPherson et al17 and 3 years in the study by Wong et al.14 The relatively short follow-up period in this study could be insufficient for the progression of NAFL to NASH.
Previous studies have shown that type 2 diabetes mellitus is one of the major driving forces of fibrosis progression.12, 18, 19 One of the clinical factors that contributed to fibrosis progression in our study was the higher presence of type 2 diabetes in progressors than in nonprogressors (30.5% vs 25%, P < 0.04). In a study involving 18,754 patients with NAFLD, of the patients who progressed, 24.7% had diabetes mellitus and 12.3% did not have diabetes mellitus.20 Studies have found that type 2 diabetes is an independent risk factor for the liver-related mortality in patients with NAFLD and that type 2 diabetes has greater liver-related mortality in patients with NASH compared with patients without NASH.19 It is important to note, despite the insignificance, there is a higher percentage of diabetics in the NASH group vs. the NAFL group (40% vs. 26.9%) at baseline in this study. The findings in this study and previous studies stress the importance of screening patients with NAFLD and type 2 diabetes for fibrosis progression. Liver damage in NASH is hypothesized to be caused by the “two-hit” hypothesis.21, 22 In patients with diabetes and NASH, insulin resistance can lead to hepatocyte lipotoxicity.23, 24, 25 The death of hepatocytes leads to activation of wound-healing responses. The responses can further lead to progressive fibrosis and increase the risk of hepatocellular carcinoma.23, 24, 25
Owing to the liver-related complications, it is important to recognize patients with advanced fibrosis. In this study, there was a noteworthy correlation between the progression of fibrosis stages and the noninvasive scoring systems such as AST/ATL ratio, NAFLD, and FIB-4. AST/ATL ratio, NAFLD, and FIB-4 were higher in fibrosis progressors, and as fibrosis progressed, these scores also increased. Previous studies have shown that NAFLD, FIB-4, and APRI score can identify patients with advanced fibrosis.26 In a meta-analysis by Singh et al,16 AST/ALT ratio was also associated more closely with fibrosis progression. In the current cohort, the APRI score did not identify the fibrosis progressors.
The limitations of the present study inherent to its retrospective nature and small sample size lead to potential selection bias. Unmeasured confounding variables may have potentially influenced the natural progression of NAFLD. Information on the portal tract was not consistently reported in a small proportion of the patients, but the length of the specimen retrieved was all ≥10 mm and was read by one of the three expert liver pathologists at our center, minimizing interobserver and intraobserver variabilities. The indications for repeat biopsy in patients with NAFLD varied. It is possible that more compliant patients may have followed up for a repeat biopsy after making recommended lifestyle choices to slower the progression of the disease. This may have altered the natural history of the disease. We used the NASH CRN fibrosis staging system score to divide patients into NAFL and NASH subgroups. While the NAS system is the most validated system presently available,27 there have been studies that demonstrated that the diagnosis of steatohepatitis based on liver biopsies may not always correlate with the NAS score.10 Interestingly, in our study, fibrosis progression was seen in patients with higher fibrosis scores at baseline. This may indicate increased disease activity in these patients, thereby increasing the rate of fibrosis progression. However, it may indicate that these patients had advanced fibrosis initially, which may have been missed by the baseline pathology analysis because of sampling variability. Some experts suggest considering bilobar liver biopsies to reduce this variability and improve assessment of disease activity and fibrosis28.
In conclusion, this study demonstrated that NAFLD has a variable histological course. An important factor that was associated with disease advancement was diabetes. Simple noninvasive scoring systems such as AST/ALT ratio, FIB-4 score, and NAFLD fibrosis score can identify patients with NAFLD at risk of fibrosis progression.
Authors' contributions
S.K.S. conceptualized and designed the study. S.G., P.S.B.P., Y.K.R., and H.K.M. collected the data. S.K.S. and Y.J. performed the statistical analysis. Y.K.R. and H.K.M. interpreted the data and wrote the initial draft of the manuscript. All other authors participated in the critical revision of the manuscript for important intellectual content.
Conflicts of interest
The authors have none to declare related to the content of the manuscript.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jceh.2019.07.002.
Appendix A. Supplementary data
The following is the supplementary data to this article:
References
- 1.Younossi Z., Anstee Q.M., Marietti M. Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nat Rev Gastroenterol Hepatol. 2018;15:11–20. doi: 10.1038/nrgastro.2017.109. [DOI] [PubMed] [Google Scholar]
- 2.Vernon G., Baranova A., Younossi Z.M. Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther. 2011;34:274–285. doi: 10.1111/j.1365-2036.2011.04724.x. [DOI] [PubMed] [Google Scholar]
- 3.Chalasani N., Younossi Z., Lavine J.E. The diagnosis and management of non-alcoholic fatty liver disease: practice Guideline by the American Association for the Study of Liver Diseases, American College of Gastroenterology, and the American Gastroenterological Association. Hepatology (Baltimore, MD) 2012;55:2005–2023. doi: 10.1002/hep.25762. [DOI] [PubMed] [Google Scholar]
- 4.Stal P. Liver fibrosis in non-alcoholic fatty liver disease - diagnostic challenge with prognostic significance. World J Gastroenterol. 2015;21:11077–11087. doi: 10.3748/wjg.v21.i39.11077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ekstedt M., Hagstrom H., Nasr P. Fibrosis stage is the strongest predictor for disease-specific mortality in NAFLD after up to 33 years of follow-up. Hepatology (Baltimore, MD) 2015;61:1547–1554. doi: 10.1002/hep.27368. [DOI] [PubMed] [Google Scholar]
- 6.Benedict M., Zhang X. Non-alcoholic fatty liver disease: an expanded review. World J Hepatol. 2017;9:715–732. doi: 10.4254/wjh.v9.i16.715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Noureddin M., Vipani A., Bresee C. NASH leading cause of liver transplant in women: updated analysis of indications for liver transplant and ethnic and gender variances. Am J Gastroenterol. 2018;113:1649–1659. doi: 10.1038/s41395-018-0088-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mohamed R., Nabih M., ElShobaky M., Khattab H. The value of noninvasive scoring systems for the diagnosis of advanced fibrosis in Egyptian patients with nonalcoholic fatty liver disease. Egypt J Intern Med. 2014;26:162–169. [Google Scholar]
- 9.Kleiner D.E., Brunt E.M., Van Natta M. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology (Baltimore, MD) 2005;41:1313–1321. doi: 10.1002/hep.20701. [DOI] [PubMed] [Google Scholar]
- 10.Brunt E.M., Kleiner D.E., Wilson L.A., Belt P., Neuschwander-Tetri B.A., Network N.C.R. Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology (Baltimore, MD) 2011;53:810–820. doi: 10.1002/hep.24127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Teli M.R., James O.F., Burt A.D., Bennett M.K., Day C.P. The natural history of nonalcoholic fatty liver: a follow-up study. Hepatology (Baltimore, MD) 1995;22:1714–1719. [PubMed] [Google Scholar]
- 12.Adams L.A., Sanderson S., Lindor K.D., Angulo P. The histological course of nonalcoholic fatty liver disease: a longitudinal study of 103 patients with sequential liver biopsies. J Hepatol. 2005;42:132–138. doi: 10.1016/j.jhep.2004.09.012. [DOI] [PubMed] [Google Scholar]
- 13.Dam-Larsen S., Franzmann M., Andersen I.B. Long term prognosis of fatty liver: risk of chronic liver disease and death. Gut. 2004;53:750–755. doi: 10.1136/gut.2003.019984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wong V.W., Wong G.L., Choi P.C. Disease progression of non-alcoholic fatty liver disease: a prospective study with paired liver biopsies at 3 years. Gut. 2010;59:969–974. doi: 10.1136/gut.2009.205088. [DOI] [PubMed] [Google Scholar]
- 15.Pais R., Charlotte F., Fedchuk L. A systematic review of follow-up biopsies reveals disease progression in patients with non-alcoholic fatty liver. J Hepatol. 2013;59:550–556. doi: 10.1016/j.jhep.2013.04.027. [DOI] [PubMed] [Google Scholar]
- 16.Singh S., Allen A.M., Wang Z., Prokop L.J., Murad M.H., Loomba R. Fibrosis progression in nonalcoholic fatty liver vs nonalcoholic steatohepatitis: a systematic review and meta-analysis of paired-biopsy studies. Clin Gastroenterol Hepatol. 2015;13:643–654. doi: 10.1016/j.cgh.2014.04.014. e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McPherson S., Hardy T., Henderson E., Burt A.D., Day C.P., Anstee Q.M. Evidence of NAFLD progression from steatosis to fibrosing-steatohepatitis using paired biopsies: implications for prognosis and clinical management. J Hepatol. 2015;62:1148–1155. doi: 10.1016/j.jhep.2014.11.034. [DOI] [PubMed] [Google Scholar]
- 18.Loomba R., Abraham M., Unalp A. Association between diabetes, family history of diabetes, and risk of nonalcoholic steatohepatitis and fibrosis. Hepatology (Baltimore, MD) 2012;56:943–951. doi: 10.1002/hep.25772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rafiq N., Bai C., Fang Y. Long-term follow-up of patients with nonalcoholic fatty liver. Clin Gastroenterol Hepatol. 2009;7:234–238. doi: 10.1016/j.cgh.2008.11.005. [DOI] [PubMed] [Google Scholar]
- 20.Simeone J.C., Bae J.P., Hoogwerf B.J. Clinical course of nonalcoholic fatty liver disease: an assessment of severity, progression, and outcomes. Clin Epidemiol. 2017;9:679–688. doi: 10.2147/clep.S144368. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Day C.P. Non-alcoholic steatohepatitis (NASH): where are we now and where are we going? Gut. 2002;50:585–588. doi: 10.1136/gut.50.5.585. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Day C.P., James O.F.W. Steatohepatitis: a tale of two “hits”? Gastroenterology. 1998;114:842–845. doi: 10.1016/S0016-5085(98)70599-2. [DOI] [PubMed] [Google Scholar]
- 23.Machado M.V., Diehl A.M. Pathogenesis of nonalcoholic steatohepatitis. Gastroenterology. 2016;150:1769–1777. doi: 10.1053/j.gastro.2016.02.066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Angulo P., Machado M.V., Diehl A.M. Fibrosis in nonalcoholic Fatty liver disease: mechanisms and clinical implications. Semin Liver Dis. 2015;35:132–145. doi: 10.1055/s-0035-1550065. [DOI] [PubMed] [Google Scholar]
- 25.Machado M.V., Diehl A.M. Liver renewal: detecting misrepair and optimizing regeneration. Mayo Clin Proc. 2014;89:120–130. doi: 10.1016/j.mayocp.2013.10.009. [DOI] [PubMed] [Google Scholar]
- 26.Siddiqui M.S., Yamada G., Vuppalanchi R. Diagnostic accuracy of noninvasive fibrosis models to detect change in fibrosis stage. Clin Gastroenterol Hepatol. 2019 doi: 10.1016/j.cgh.2018.12.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sanyal A.J., Brunt E.M., Kleiner D.E. Endpoints and clinical trial design for nonalcoholic steatohepatitis. Hepatology (Baltimore, Md) 2011;54:344–353. doi: 10.1002/hep.24376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Khurana S., Butt W., Khara H.S. Bi-lobar liver biopsy via EUS enhances the assessment of disease severity in patients with non-alcoholic steatohepatitis. Hepatol Int. 2019;13:323–329. doi: 10.1007/s12072-019-09945-4. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.