Nghiên cứu đặc điểm lâm sàng, một số xét nghiệm, siêu âm và mô bệnh học bệnh gan nhiễm mỡ tt TIẾNG ANH

MINISTRY OF EDUCATION AND TRAINING
MINISTRY OF DEFENCE
108 INSTITUTE OF CLINICAL MEDICAL AND PHARMACETICAL SCIENCES

VU THI THU TRANG

RESEARCH ON CLINICAL FEATURES, SOME TESTS,
ULTRASOUND AND HISTOPATHOLOGY
OF FATTY LIVER DISEASE
Speciality: Gastroenterology
Code: 62.72.01.43

ABSTRACT OF MEDICAL PHD THESIS

HA NOI - 2019


THE THESIS WAS DONE IN: 108 INSTITUTE OF CLINICAL
MEDICAL AND PHARMACEUTICAL SCIENCES

Supervisor:
1. Prof. PhD. Trinh Tuan Dung
2. PhD. Duong Minh Thang

Reviewer:
1.
2.
3.

This thesis will be presented at Institute Council at:
108 Institute of Clinical Medical and Pharmaceutical Sciences

Day

Month

Year

The thesis can be found at:

1.
2.

National Library of Vietnam
Library of 108 Institute of Clinical Medical and
Pharmaceutical Sciences


LIST OF PUBLISHED ARTICLE RELATING TO THESIS
1. Vu Thi Thu Trang, Trinh Tuan Dung, Nguyen Tien Thinh, Duong
Minh Thang (2018), "Clinical and laboratory characteristics of fatty
liver disease". Journal of 108 - Clinical Medecine and Pharmacy ,
13:247-251.
2. Vu Thi Thu Trang, Trinh Tuan Dung, Nguyen Tien Thinh, Duong
Minh Thang (2018), "Histopathological characteristics of fatty liver
disease". Journal of 108 - Clinical Medecine and Pharmacy, 13:258263.

3. Vu Thi Thu Trang, Trinh Tuan Dung, Nguyen Tien Thinh,
Duong Minh Thang (2019), "Relationship between clinical,
laboratory and fibrosis and value of non-invasive diagnostic
methods for liver fibrosis in fatty liver disease". Journal of 108 Clinical Medecine and Pharmacy, 14:57-62.

4
INTRODUCTION
Fatty liver disease (FLD) is an abnormal accumulation of
fat (mainly triglycerides) in liver cells due to many causes such as
alcohol, metabolic disorders, nutritional disorders, poisoning,
drugs, hepatitis virus ... The disease progresses silently from the
steatosis stage to steatohepatitis and eventually to cirrhosis. In
fact, alcoholic fatty liver disease (AFLD) and nonalcoholic fatty
liver disease (NAFLD) are common. AFLD is all cases of alcohol
abuse, while NAFLD includes metabolic and nutritional liver
diseases that are often the result of insulin resistance and fat
According to the World Health Organization statistics, the
rate of fatty liver disease worldwide varies from 4% to 46%
depending on the region and region. Over the years, non-invasive
diagnostic methods have been constantly researched and
developed to replace liver biopsies. However, the gold standard
for diagnosing steatosis is still a liver biopsy and
histopathological examination. A liver biopsy not only helps
diagnose fatty liver disease, but more importantly can accurately
diagnose the level and stage of the disease, help predict and
evaluate the effectiveness of treatment of fatty liver disease.
There are many researches on fatty liver disease in the
world that show the serious consequences of the disease for
individuals as well as the whole society. In Vietnam, fatty liver
disease is also on the rise and is currently a topical issue of the
health sector and is of great concern to society. However, so far
there is no specific statistics on fatty liver status and there have
not been many studies on fatty liver disease in Vietnam,
especially studies with liver biopsy as gold standard in the

diagnosis of fatty liver disease. Therefore, we conduct this study
for two main objectives:
1. Analyze clinical features, some tests, ultrasound and
histopathology in fatty liver disease.
2. Assessment the relationship between clinical, laboratory,
ultrasound and histopathology and the value of some non-


5
invasive diagnostic methods in fatty liver disease.
Thesis layout
The thesis has 136 pages, including: Introduction (2 pages),
Chapter 1: Overview (34 pages), Chapter 2: Subjects and
methods (22 pages), Chapter 3: Results (40 pages), Chapter 4:
Discussion (35 pages), Conclusion (2 pages), Recommendations
(1 page). The thesis has 51 tables, 16 Charts, 19 Pictures. The
thesis has 129 References (8 vietnamese references and 121
english references) and 2 appendices.
The new main contributions of the thesis
This is the first study to evaluate the detailed histopathology
of fatty liver disease in Vietnam. The study has given relatively
comprehensive results on fatty liver disease including clinical
features, tests, ultrasound and histopathology of fatty liver disease
and the difference between AFLD and NAFLD. The thesis also
assessed the relationship between clinical tests, ultrasound and
the degree of histopathological damage in fatty liver disease. In
particular, the study has shown that the value of some noninvasive diagnostic methods in fatty liver disease such as APRI,
FIB-4, Forns has good value for diagnosing liver fibrosis in fatty
liver disease and The ANI index is a very good indicator for the
differential diagnosis of alcoholic and nonalcoholic fatty liver.

These are valuable contributions both theoretically and
practically in the diagnosis and treatment of steatosis.


6
CHAPTER I. OVERVIEW
1.1. Epidemiology of fatty liver disease
1.1.1. AFLD

Alcohol causes more than 200 different diseases and
injuries on the body. Alcohol is the first cause of liver disease in
the West and the second cause in Asia after viral hepatitis.
1.1.2. NAFLD

According to the World Gastroenterology Organisation, the
American association for the study of the liver diseases and the
European association for the study of the liver, NAFLD is the
most common cause of liver diseases worldwide. The prevalence
of NAFLD worldwide varies 4 to 46%, while the prevalence of
nonalcoholic steatohepatitis is lower limit of 3-5%.
1.2. Definition, causes, classification, pathogenesis of fatty
liver disease
1.2.1. Definition

Fatty liver is characterized by excessive accumulation of fat
(mainly triglycerides) in liver cells. FLD is defined as having
more than 5% of steatosis hepatocytes diagnosed by
histopathology or magnetic resonance imaging.
1.2.2. Causes

* Causes of chronic FLD: Alcoholism, obesity, diabetes,
dyslipidemia, reversible surgery - jejunum, protein - energy
deficiency, parenteral nutrition, rapid weight loss, genetic
disorders of fatty acid oxidation in mitochondria, other liver
diseases ...
* Causes of acute FLD: Alcohol poisoning, FLD in
pregnancy, Reye's syndrome, Jamaican vomiting, Wolman
disease, toxins ...
1.2.3. Classification

1.2.3.1. Classification by histopathology: macrovesicular
steatosis, microvesicular steatosis, mixture of large and small


7
vacuoles steatosis.
1.2.3.2. Classification by cause: alcoholic fatty liver, nonalcoholic
fatty liver, secondary hepatic steatosis.
1.2.4. Pathogenesis
The accumulation of fat in the liver essentially results from four
pathogenetic processes:
- Increased uptake of free fatty acids (from food or body fat) via
the portal vein.
- Increased synthesis of free fatty acids in the liver (from glucose
or acetate).
- Decreasedβ-oxidation of free fatty acids (above all in the
mitochondria).
- Decreased synthesis or secretion of lipoproteins (very low
density
lipoproteins, VLDL), the principal route for elimination of lipids

from the liver.
1.3. Clinical
The clinical manifestations of FLD depend on the cause and
stage of the disease. The majority of steatosis cases have no
clinical or non-specific symptoms.
1.4. Biological markers for diagnosing FLD
1.4.1. Biological markers of hepatic steatosis
1.4.1.1. FLI (Fatty liver index)
1.4.1.2. HSI (hepatic steatosis index)
1.4.1.3. LAP (lipid accumulation product)

1.4.2. Biological markers identifying steatohepatitis and
cirrhosis
1.4.2.1. Liver enzymes AST, ALT,GGT

1.4.2.2. The APRI index
1.4.2.3. The Forns index
1.4.2.4. The FIB-4 index
1.4.3. The ANI index distinguishes between AFLD and
NAFLD
The ANI index is simple, convenient, and can be used in clinical


8
practice to distinguish between AFLD and NAFLD.
ANI = -58,5 + 0,637 (MCV) + 3,91 (AST/ALT) - 0,406 (BMI) + 6,35
(male)
1.5. Image diagnostic methods
1.5.1. Ultrasound
Ultrasound is a simple, convenient, inexpensive, most

commonly used method for screening for FLD and is often the
first choice to assess fatty liver. Reports of sensitivity and
specificity of ultrasound to detect all stages of hepatic steatosis
are 60-94% and 66-95%, respectively. However, ultrasound is
only good for moderate to severe FLD (> 30% of fat cells are fat).
1.5.2. Computed tomography - CT
Computed tomography is highly sensitive and specific in
the diagnosis of steatosis, with moderate and severe steatosis
(≥30% histologically), CT scan has a sensitivity of 84% and a
specificity of 100%. However, CT scans are less sensitive in
cases of mild FLD below 30% and the specificity of diagnosing
fatty liver is also affected by factors such as fibrosis,
inflammation and edema, therefore CT scans are not distinguish
between simple fatty liver and steatohepatitis.
1.5.3. Magnetic resonance imaging - MRI
MRI scans are a more accurate and relatively superior
technique than ultrasound and CT scans to detect mild cases of
fatty liver. MRI can be used to quantify fat in liver parenchyma.
However, MRI scans are still difficult to distinguish from
inflammation and fibrosis, however, due to the high cost of MRI
has not been widely used clinically to diagnose fatty liver.
1.5.4. Techniques liver elastography
The general value of methods of measuring liver elasticity
in FLD is to allow the assessment of liver fibrosis. Fibroscan also
allows the diagnosis of hepatic steatosis but does not accurately
assess the stages and types of lesions in FLD.


9
1.6. Histology of FLD

1.6.1. Steatosis: There are two types of steatosis, macrovesicular
steatosis and microvesicular steatosis.
1.6.1.1. Macrovesicular steatosis

Marcovesicular steatosis is caused by the deposition of fat
in the cytoplasm of liver cells that form large fat particles pushing
the nucleus to one side.
1.6.1.2. Microvesicular steatosis

Microvesicular steatosis is often associated in the
metabolism of fat in the liver cells, causing the mitochondria to
transform into many small drop-fat located in the cytoplasm of
the hepatocytes surrounding the nucleus of the cell, not repelling
the nucleus.
1.6.2. Hepatocellular lesions

1.6.2.1. Hepatocyte balloning
1.6.2.2. Acidophil bodies (Apoptotic hepatocyte)
1.6.2.3. Necrosis of hepatocytes
1.6.3. Lobular and portal inflamation
1.6.4. Fibrosis
1.6.5. Other lesions in FLD
1.6.5.1. Mallory-Denk bodies

1.6.5.2. Nuclear vacuolation (Glycogenated nuclei)
1.6.5.3. Megamitochondria
1.6.6. Diagnose stage and degree of steatosis: by NAS score
and FLIP algorithm. There are 3 stages: simple steatosis,
steatohepatitis, cirrhosis
CHAPTER 2. SUBJECTS AND METHODS

2.1. Subjects

102 patients with FLD were diagnosed and treated at 108
Military Central Hospital from 12/2013 to 12/2016.
2.1.1. Subject selection

Patients ≥ 18 years of age, diagnosed by histopathology of
fatty liver regardless of alcohol, non-alcohol, or other causes such


10
as hepatitis B, C, drugs ...
Criteria for diagnosis of FLD:
Diagnosis of FLD by histology when ≥ 5% of hepatocytes
are steatosis (count the number of hepatocytes per 5 domains at
magnification 400 times, divided by the average number).
2.1.2. Exclusive criteria

- Patients do not agree to participate in the study.
- Pregnant and and nursing mothers.
- Patients with serious internal diseases such as kidney
failure, severe heart failure, respiratory failure, severe infection,
decompensated cirrhosis, liver cancer, liver abscess, biliary tract
infection…
- An unqualified liver biopsy: less than 4 portals, less than
1.5 cm in length.
2.2. Methods
2.2.1. Research design: Cross-sectional descriptive study
2.2.2. Cỡ mẫu: Convenience sample, n=102.
2.2.3. Research criteria

2.2.3.1. Cause: 3 groups: alcoholic fatty liver, non-alcoholic fatty
liver, fatty liver from many causes / unidentified cause.
2.2.3.2. Clinical criterias

* Age, gender, BMI, waist circumference.
* History: Alcohol abuse, diabetes, hypertension, dyslipidemia,
metabolic syndrome, hepatitis B, C ...
* Functional and physical symptoms: Fatigue, abdominal
distension, stool disorders, right upper abdominal pain, spider
angioma, jaundice, hepatomegaly, asymptomatic.
2.2.3.3. Hematology and blood biochemistry
- Hematology: RBC count, WBC count, platelet count, Hb, MCV,
Prothrombin ratio.
- Biochemistry: Urea, creatinine, total bilirubin, total protein,
albumin, fasting blood glucose, AST, ALT, GGT, total
cholessterol, triglycerides, LDL-Cholesterol, HDL-Cholesterol.


11
2.2.3.4. Liver ultrasound criteria

* Grade of steatosis: 3 grades (I, II, III)
* Ultrasound lesions: Hepatomegaly, diffuse steatosis,
pseudotumor (hypoechoic foci, hyperechoic foci, echo mixture
foci).
2.2.3.5. Histology criteria

* The stages of FLD: 3 stages (simple fatty liver, steatohepatitis,
cirrhosis).
* Steatosis:

- Grade of steatosis: 3 grades (1, 2, 3)
- Classification of steatosis: macrovesicular steatosis,
microvesicular steatosis, and mixed steatosis (include
macrovesicular steatosis and microvesicular steatosis ).
- Location of steatosis: zone 3, zone 1, diffuse.
* Inflammation
- Lobular inflammation: 3 grades (mild, moderate, severe).
- Portal inflammation: 3 grades (mild, moderate, severe)
- Cells of inflammation: lympho and mixed cells
- Grade of steatohepatitis: NAS score.
* The damage to liver cells:
hepatocyte ballooning,
hepatocellular necrosis, acidophil bodies Mallory–Denk bodies,
nuclear vacuolation, megamitochondria.
* Other lesions: Large lipogranuloma, microgranulomas, ductular
reaction, hepatocellular dysplasia
* Fibrosis: Evaluation by 3 staining methods (H.E. staining,
Trichrome Masson staining, Vimentin staining).
- Locations of liver fibrosis
- Stade of liver fibrosis: Evaluation by 2 score (Metavir and
NAS).
2.2.3.6. Non-invasive diagnosis of cause, steatosis and fibrosis

- ANI: Differential diagnosis of ALD and NAFLD
- Test of steatosis: HSI, FLI, LAP.
- Test of fibrosis: APRI, FIB-4, Forns.
2.2.4. Equipment


12

Modern testing machines, imaging and histopathology
diagnostic facilities are routinely used in the 108 hospital, biopsy
guns (fast-gun), biopsy needles (fast-cut).
2.2.5. Steps of research process
2.2.5.1. Clinical examination
2.2.5.2. Hematological and biochemical blood test
2.2.5.3. Ultrasound
2.2.5.4. Biopcy
A liver biopsy is performed using a Fast-Gun biopsy device
and a Fast-Cut 16G liver biopsy under the guidance of ultrasound
at the Department of Gastroenterology (A3), 108 Military Central
Hospital following the procedures of the Ministry of Health.
* Indication: Conduct liver biopsy for patients with suspected
steatohepatitis, fatty liver patients on ultrasound image of
combination lesions needed biopsy for definitive diagnosis. Cases
of coordinated lesions will take two pieces of liver tissue: one
piece of normal liver tissue and one piece at the location of
suspected lesions.
* Contraindications: According to the regulations of the Ministry
of Health.
2.2.4.5. Histopathological test: performed at the Department of
Pathological Anatomy - the 108 Military Central Hospital
* Fixing and handling: liver biopsy pieces are fixed in the biopsy
room with 10% neutral buffer formol and transferred to the
Department of Pathological Anatomy within 24 hours.
* Template staining: By 3 different techniques: HematoxylinEosin (HE) staining, Trichrome Masson staining and
immunohistochemistry staining with Vimentin antibody
(Vimentin staining).
2.2.6. Statistical analysis: By medical statistical methods, using
SPSS 21.0 software.

13
2.3. Research ethics: In accordance with the regulations of the
108 Military Central Hospital and the 108 Institute of Clinical
Medical and Pharmaceutical Ssiences
CHAPTER III. RESULTS
3.1. Clinical characteristics of FLD
* Cause: NAFLD 42,2%; AFL 33,3%, fatty liver from many
causes / unidentified cause 24,5%.
* Stage: steatohepatitis 83,3%, simple fatty liver 11,8%, cirrhosis
4,6%.
* Age, gender: The average age of patients with FLD is 49.77 ±
11.89. The ratio of male to female is 2:1.


14
Table 3.3. Body mass index and waist circumference of FLD
patients
Index
n
%
Waist
Normal
49
48,0
circumference
Increase
53
52,0

(cm)
Normal
33
32,3
Overweight
36
35,3
BMI (kg/m2)
Obesity
33
32,4
Average BMI
23,99 ± 2,04
Comment: waist circumference was 52%, overweight was 35.3%,
obesity was 32.4%, average BMI was 23.99 ± 2.04.
Table 3.4. Body indicators of AFLD and NAFLD patients
Group
AFLD
NAFLD
p
Index
(n = 34)
(n = 43)
Increase WC
3 (8,8)
41 (95,3)
BMI normal
21 (61,8)
1 (2,3)
Overweight

13 (38,2)
14 (32,6)
< 0,05
Obesity
0
28 (65,1)
Average BMI
22,52 ± 1,51
25,47 ± 1,52
Comment: NAFLD had average BMI and higher rate of
overweight and obesity than AFLD, the difference was
statistically significant.
Table 3.6. Clinical symptoms are seen in patients with FLD (n =
102)
Symptoms
n
%
Fatigue
42
41,2
Abdominal distension
29
28,4
Stool disorders
30
29,4
Right upper abdominal pain
27
26,5
Hepatomegaly

15
14,7
Spider angioma
13
12,7
Jaundice
3
2,9


15
Asymptomatic
35
34,3
Comment: 34.3% asymptomatic, fatigue 41.2%; abdominal
distension 28.4%; stool disorders 29.4%; right upper abdominal
pain 26.5%.
* Steatohepatitis and cirrhosis have more clinical symptoms than
simple fatty liver.
* Clinical symptoms are more common in AFLD than in NAFLD
group.
3.2. Laboratory characteristics of patients with FLD
* Routine hematological tests are almost normal.
* AFLD had an average MCV (95.94 ± 10.5fl) and an increase
MCV ( 41.1%) higher than NAFLD (85.7 ± 5.33 fl; 2.3%), p
<0.001.
Table 3.11. Serum hepatic enzyme tests in patients with FLD
Index
AST (U/l)
ALT (U/l)

GGT (U/l)
Value
n
35
40
58
increas
%
34,5
39,2
56,9
e
Range
13 - 430
8 - 443
8 - 1866
Median
30
33
62
186,9 ±
52,57 ± 65,39
48,4 ± 55,5
343,59
Comment: Liver enzymes fluctuate in a large range. Increase of
GGT is 56.9%, increase of AST is 34.5%, increase of ALT is
39.2%.
* The group of simple steatosis had almost no increase in liver
enzymes, the group of steatohepatitis and cirrhosis had 38.9%
increase in serum AST; 44.4% increased serum ALT; 62.2%

increased serum GGT.
* AFLD had higher rate of increase and activity of AST and GGT
than NAFLD group, the difference was statistically significant
with p <0.05.
* 24.7% of patients with FLD had hyperglycemia, 16.8% had
fasting blood sugar disorders. 67.6% of patients had


16
dyslipidemia, of which 41.8% had hypercholesterolemia; 50.5%
increased blood triglycerides; 31.9% increase LDL blood; 22.2%
decreased HDL blood.
* The group of steatohepatitis had a higher rate of dyslipidemia
and increased triglycerides than the group of simple fatty liver.
NAFLD group had higher LDL-cholesterol increase rate than
AFLD group, the difference was statistically significant p <0.05.
3.3. Features of liver ultrasound of patients with FLD
Table 3.16. Features of liver steatosis on ultrasound
Ultrasound
n
Hepatomegaly
11
Diffuse and evenly steatosis
43
Hypoechoic
46
Diffuse
steatosis
Hyperechoic
12

with pseudo-tumor
Echo mixture
1
image
Total
59
Comment: 10.8% of patients with FLD had hepatomegaly on
ultrasound. The rate of pseudotumor is 58.9% of which 45.1% is
hypoechoic foci.
3.4. Histopathological characteristics of FLD
Table 3.17. Features of liver steatosis
Tổn thương
n
%
Grade I
42
41,2
Grade of steatosis
Grade II
45
44,1
Grade III
15
14,7
Macrovesicul
63,7
65
ar
Classification of
Microvesicul

2,9
steatosis
3
ar
Mixture
34
33,4
Zone 1
3
2,9
Location of
Zone 3
23
22,6
steatosis
Diffuse
76
74,5


17
Comment: There are 41.2% steatosis grade I and 44.1% grade II.
Marcrovesicular is 63.7% and diffuse steatosis is 74.5%.
Table 3.19. Lobular inflammation characteristics of patients with
FLD
Lesion
n
None
5
Mild

65
Lobular
inflammation
Moderate
26
Severe
6
Comment: lobular inflammation was 95%, the majority is mild
(65%).
Table 3.20. Portal inflammation characteristics of patients with
FLD
Lesion
n
%
None
16
15,7
Portal
Mild
47
46,1
inflammatio
Moderate
29
28,4
n
Severe
10
9,8
Lympho

64
62,7
Inflamed
Mixture
with
cells
38
37,3
neutrophils
Comment: Portal inflammation accounts for 84.3%, mostly in
mild (46.1%) and moderate (28.4%).
Table 3.22. Characteristics of hepatocellular lesions of FLD
Lesion
n
%
None
8
7,8
Hepatocyte
Few
44
43,2
balloning
Many
50
49,0
None
43
42,2
Necrosis of

Few
49
48,0
hepatocytes
Many
10
9,8
Acidophil
None
10
9,8
bodies
Few
76
74,5

6
2


18
Many
16
15,7
Microgranulomas
83
81,4
Large lipogranuloma
17
16,7

Nuclear vacuolation
39
38,2
Mallory-Denk bodies
36
35,3
Megamitochondria
17
16,7
Ductular reaction
6
5,9
Comment: The most common lesions are hepatocyte ballooning,
acidophil bodies, and necrosis of hepatocytes.
* Steatohepatitis and cirrhosis have more hepatocyte ballooning,
acidophil bodies, and hepatocellular necrosis than simple fatty
liver.
* Hepatocellular necrosis, Mallory-Denk bodies, mitochondria in
AFLD group are higher than in NAFLD group; Nuclear
vacuolations were found in NAFLD (51.2%) than AFLD
(23.5%), the difference was statistically significant p <0.05.
Table 3.25. Characteristics of liver fibrosis by staining method
Staining
Trichrome
HE
Vimentin
method
Masson
n (%)
n (%)

Fibrosis
n (%)
Perisinusoidal
14 (13,7)
77 (75,5)
94 (92,1)
Pericellular
4 (3,9)
32 (31,1)
97 (95,1)
Portal
40 (39,2)
71 (69,6)
82 (80,4)
Septa fibrosis
13 (12,7)
22 (21,6)
23 (22,5)
Bridging fibrosis
8 (7,8)
13 (12,7)
15 (14,7)
Cirrhosis
5 (4,9)
5 (4,9)
5 (4,9)
Comment: Trichrome Masson and Vimentin staining revealed
higher fibrosis rates than HE staining. Most patients with FLD
already have fibrosis
Table 3.28. Fibrosis characteristics of AFLD and NAFLD

Group
AFLD
NAFLD
(n = 34)
(n = 43)
p
Fibrosis
n (%)
n (%)
Perisinusoidal
33 (97,1)
37 (86)
> 0,05


19
Pericellular
34 (100)
38 (88,4)
> 0,05
Portal
32 (94,1)
27 (62,8)
< 0,001
Septa fibrosis
11 (32,4)
6 (14,0)
> 0,05
Bridging fibrosis
7 (20,6)

3 (7,0)
> 0,05
Cirrhosis
4 (11,8)
0
Comment: AFLD had a higher incidence of portal fibrosis
(94.1%) than NAFLD (62.8%), the difference was statistically
significant, p <0.001.
* On the NAS, the AFLD group had a higher moderate-severe
fibrosis than the NAFLD group, the difference was statistically
significant with p = 0.001.
3.5. Relationship between clinical, laboratory, ultrasound and
histopathology of FLD
* There was a positive correlation between waist circumference
and BMI with the grade of steatosis in NAFLD group (r = 0.43
and 0.35; p <0.05).
* There was a positive correlation between AST with hepatic
steatosis, inflammation and fibrosis (r: 0.289; 0.373; 0.271; p
<0.05), and ALT with the grade of steatohepatitis (r = 0.308 , p
<0.05). There was a positive correlation between GGT and the
grade of steatohepatitis and the stading fibrosis (r = 0.237 and
0.371; p <0.05).
* There was a positive correlation between the grade of steatosis
and the grade of steatohepatitis (r = 0.82; p <0.05), between the
grade of steatosis and the stading fibrosis (r = 0.283; p <0.05). ),
between the grade of hepatitis and the stading fibrosis (r = 0.421;
p <0.05).
* There is poor correlation between ultrasound and
histopathology in diagnosing the grade of steatosis with Kappa
coefficient = 0.152; p <0.05

3.6. The value of non-invasive diagnostic methods for FLD


20
Figure 3.14. ROC
curve of ANI index
in
AFLD
and
NAFLD diagnosis

Comment: ANI index has good value for differential diagnosis of
AFLD and NAFLD with AUC = 0.96 and p <0.001.

Figure 3.15. ROC curve of non-invasive tests in the differential
diagnosis of liver fibrosis F3-F4 with liver fibrosis F0-F2
Comment: APRI, FIB-4, Forns are all good values for the
differential diagnosis of severe liver fibrosis (F3-F4) with mild
and moderate liver fibrosis (F0-F2) with AUROC> 0.8 and p
<0.001.


21

Figure 3.15. ROC curve of non-invasive tests in the differential
diagnosis of liver fibrosis F3-F4 with liver fibrosis F0-F2
Comment: The APRI, FIB-4 and Forns tests are both moderately
valuable to distinguish moderate and severe fibrosis (F2-F4) from
mild fibrosis (F0-F1) with AUROC ≥ 0.7. and p <0.05.
CHAPTER 4. DISCUSSIONS

4.1. Clinical characteristics of patients with FLD
The mean age of patients with FLD in our study was 49.47
± 11.89 years. The ratio of male to female is 2: 1.
The rate of overweight was 35.3%, the obesity rate was
32.4%, and the increase in waist circumference accounted for
52%. 95.3% NAFLD group had an increase in waist
circumference, 65.1% were overweight, 32.6% were overweight,
while AFLD group was mostly normal. The average BMI of
NAFLD (25.46 ± 1.52)íis higher than AFLD (22.53 ± 1.53), the
difference was statistically significant p <0.05.
In this study, 34.3% of the subjects were asymptoms. Some
common functional symptoms are fatigue (41.2%), distribution
(28.4%), stool disorders (29.4%), or lower right flank pain
(26.5%). ). There are very few physical symptoms, only
hepatomegaly accounts for 14.7%. Steatohepatitis is more
clinically manifest than simple fatty liver. AFLD has more


22
clinical manifestations than NAFLD.
4.2. Laboratory characteristics of patients with FLD
Our research results show that most patients with fatty liver
have hematological tests and normal liver function. AFLD
patients had an average MCV of 95.94 ± 10.5 fL and 41.1% had
an increase in MCV, higher than NAFLD group (average MCV of
85.7 ± 5.33 fL and 2.3% increase in MCV), the difference is
statistically significant with p <0.05.
Regarding serum liver enzymes, our research shows that
serum liver enzyme values fluctuate in a large range. There was
34.5% of AST increased, 39.2% of ALT increased and 56.9% of

GGT increased serum. In which, steatohepatitis has a higher
serum liver enzymes than simple fatty liver. AFLD has a higher
AST and GGT than NAFLD. There was no difference in ALT
between AFLD and NAFLD.
In this study, 24.8% of patients with FLD had fasting
hyperglycemia, 16.8% of fasting blood sugar disorders, average
of level fasting blood glucose at a high of 6.61 ± 2.44 mmol / l.
There was no difference in fasting blood sugar disorders between
AFLD and NAFLD.
The prevalence of dyslipidemia in patients with FLD in the
study was 67.6%, of which 41.8% increased total cholesterol,
50.5% increased triglycerides, 31.9% increased LDL-cholesterol,
22.2% reduce HDL-cholesterol. The rate of dyslipidemia and
hypertriglyceridemia of steatohepatitis is higher than that of
simple fatty liver. There was no difference in the rate of
dyslipidemia between AFLD and NAFLD, but NAFLD had
higher LDL-cho increase rate than AFLD group.
4.3. Liver ultrasound characteristics of patients with FLD
Regarding the ultrasound characteristics, our research
results show that only 10.8% of patients with FLD have
hepatomegaly. There are 42.2% of steatosis diffuse evenly, 57.8%
of diffuse steatosis with image of pseudotumor, in which 45.1%
of hypoechoic mass, 11.8% of hyperechoic mass, image of


23
mixture echo mass only 1 patient accounts for 1%.
4.4. Histopathological characteristics of patients with FLD
4.4.1. Steatosis

In this study, the majority of cases with FLD were
macrovesicular steatosis (63.7%) and a mixture of steatosis
(33.3%), microvesicular steatosis is only 2.9%. The majority of
cases were steatosis grade 1 (41.2%) and grade 2 (44.1%).
Diffuse steatosis accounting for 74.5%, steatosis in zone 3 is
22.5%, only 3 cases of steatosis in zone 1 (periportal). There was
no difference of steatosis characteristics between the AFLD and
the NAFLD.
4.4.2. Inflammation

Research results show that the majority of patients have had
lobular inflammation and portal inflammation. Mild lobular
inflammation was 63.7%, average was 25.5%, severity was 5.9%
and 4.9% had no lobular inflammation. There was no difference
in the grade of lobular inflammation between the AFLD and the
NAFLD. The mild portal inflammation is 46.1%, moderate is
28.4%, only 9.8% is severe. The AFLD has a more severe portal
inflammation than the NAFLD.
4.4.3. Features of hepatocellular lesion

Liver cell damage in FLD is also quite diverse. The most
common lesions in this study were hepatocyte ballooning 92.2%,
acidophil bodies 90.2%, microgranulomas 81.4%, hepatocellular
necrosis 57.8%, most of them are mild hepatocellular necrosis.
Other less common injuries were nuclear vacuolation 38.2%,
Mallory-Denk bodies 35.3%, megamitochondria 16.7%, large
lipogranlulomas 16.7%, ductular reaction 5.9%. No lesions of
biliary obstruction or hepatocellular dysplasia. AFLD group had
more hepatocellular necrosis, more Mallory-Denk bodies, more
megamitochondria, less nuclear vacuolation than NAFLD group,

the difference is statistically significant.


24
4.4.4. Features fibrosis

Our research results show that, in FLD, Trichrome Masson
and Vimentin staining methods were found to have higher fibrosis
than HE staining method with both scoring systems of liver
fibrosis (NAS and Metavir). With these staining methods, most
patients in our study were diagnosed with fibrosis at different
stages. The most common fibrosis stage is pericellular fibrosis
95.1%, perisinusoidal fibrosis 92.1%, portal and periportal
fibrosis 80.4%. The severity of liver fibrosis was not much, septa
fibrosis was 22.5%, bridging fibrosis was 14.7% and cirrhosis
was 4.9%. There was no difference in pericellular fibrosis and
perisinusoidal fibrosis between the AFLD and the NAFLD.
However, the rate of portal and periportal fibrosis of the AFLD
(94.1%) was higher than the NAFLD (62.8%), the difference was
statistically significant with p <0.001. Assessment by NAS scale,
the AFLD group had moderate and severe fibrosis (94.1%) higher
than the NAFLD group (62.8%), the difference was statistically
significant with p <0.001.
4.5. Relationship between clinical, laboratory and
histopathology of FLD
In this study, there was no relationship between clinical
with histopathological in patients with FLD. In NAFLD group,
there was a positive correlation between BMI and waist
circumference with grade of steatosis (r was 0.35 and 0.43; p
<0.05). There was a positive correlation between serum AST with

grade of steatosis (r = 0.271; p <0.05), with the grade of
steatohepatitis (r = 0.373; p <0.05) and with stage of fibrosis (r =
0.289; p <0.05). There was a positive correlation between serum
ALT and grade of steatohepatitis (r = 0,308; p <0.05). There was
a positive correlation between serum GGT with grade of
steatohepatitis (r = 0.237; p <0.05) and with the stage of liver
fibrosis (r = 0.371; p <0.05).


25
4.6. The value of non-invasive diagnosis in FLD
The study results showed that there is poor similarity
between abdominal ultrasound and histopathology in the
diagnosis of hepatic steatosis with kappa coefficient = 0.152 and
p = 0.034. The HSI and LAP indices are not valuable in assessing
the grade of steatosis with AUROC <0.6 and p> 0.05, the FLI
index had also low value with AUROC = 0.664 and p <0.05.
Analyzing the value of the ANI index in the differential
diagnosis of AFLD and NAFLD, the results showed that the
AUROC of the ANI was 0.96 with p <0.001; the cutting point
was - 2.56, the sensitivity was 88.2%, the specificity was 90.7%.
In order to assess the stage of liver fibrosis in patients with
FLD, the study results showed that the APRI, FIB-4, Forns all
have good values in the diagnosis of cirrhosis and severe fibrosis
with AUROC > 0.8 , p < 0.05, they have moderate value in the
diagnosis of mild fibrosis with AUROC> 0.7, p <0.05 and not
have value in differential diagnosis of fibrosis and non-fibrosis.
CONCLUSIONS
1. Clinical, laboratory, ultrasound and histopathological
features of FLD

- The mean age was 49.77 ± 11.89. The ratio of male to female is
2: 1. The mean BMI was 23.99 ± 2.04.
- Alcohol abuse 43.1%, obesity 32.4%, abdominal obesity 52%,
clinical symptoms are poor, 34.3% have no symptoms.
-34.5% increased AST; 39.2% increased ALT; 56.9% increased
GGT; 41.6% had hyperglycemia/fasting blood sugar disorder;
70.4% had dyslipidemia, the majority was hypertriglyceridemia.
- Ultrasound: hepatomegaly was 15.7%, pseudotumor image was
57.8%.
- Steatohepatitis was 83.3%; simple steatosis was 11.8%;
cirrhosis was 4.9%. Steatohepatitis and cirrhosis have more