Nghiên cứu hội chứng chuyển hóa và nồng độ leptin, IL 1β huyết tương ở bệnh nhân thoái hóa khớp gối nguyên phát tt tiếng anh

.MINISTRY OF EDUCATION AND
MINISTRY OF
TRAINING
DEFENCE
VIETNAM MILITARY MEDICAL UNIVERSITY

NGUYEN THI THANH MAI

METABOLIC SYNDROME, PLASMA LEPTIN
AND IL-1β CONCENTRATIONS IN PATIENTS
WITH PRIMARY KNEE OSTEOARTHRITIS

Speciality

: INTERNAL MEDICINE

Code

: 9 72 01 07

THE SUMMARY OF THE MEDICAL DOTORAL THESIS

HANOI – 2019


THIS DOCTORAL THESIS WAS COMPLETED
AT VIETNAM MILITARY MEDICAL UNIVERSITY

Scientific Instructors:
1. A/PROF. Ph.D. Do Trung Quan
2. A/PROF. Ph.D. Dao Hung Hanh

1st Contradictor: A/PROF. Ph.D. Nguyen Thanh Thuy
2nd Contradictor: PROF. Ph.D. Vo Tam
3rd Contradictor: A/PROF. Ph.D. Doan Van De

The doctoral thesis will be reported to The Grading and
Examinations Committee hold at Vietnam Military Medical
University at

2019

Searching for the dissertation at:
-

National Library

-

Vietname Military Medical University’s library


1
INTRODUCTION
Osteoarthritis (OA) was the result of interaction between
biomechanics and chronic low-stage inflammation. Metabolic syndrome
(MS) included central obesity, dyslipidemia, impaired fasting glucose,
and hypertension in which obesity was the most important. The
prevalence of MS increased in the OA compared with the non-OA
group. Obesity was a significant risk factor for knee OA (KOA) through
mechanisms of increasing mechanical load and chronic low-stage

inflammation. Low-stage chronic inflammation associated with
adipokines (leptin) interacts with pro-inflammatory cytokines of which
interleukin1β (IL-1β) was a key role in maintaining cartilage ulcers,
substrate degradation in obesity pathogenesis mechanism caused OA.
Therefore, the thesis “Metabolic syndrome, plasma leptin and
IL-1β concentrations in patients with primary knee osteoarthritis”
was conducted with two objectives:
1. To determine the prevalence of metabolic syndrome, its
components and the relationship with the stages of primary knee
osteoarthritis.
2. To determine the relationship of plasma leptin and IL-1β
concentrations with some clinical and subclinical features in patients
with primary knee osteoarthritis.
* The scientific significance: Increasing the incidence of MS in
the OA, MS and OA were shared the pathogenesis mechanism related
to obesity, low-stage chronic inflammatory response and adipokines
(leptin) or pro-inflammatory cytokines (IL-1β).


2
* The practical significance: Described the specific value of the
prevalence of MS, plasma leptin and IL-1β concentrations in KOA and
compared with the control. It was recommended to identify a KOA
patient with MS accompanying control of two problems simultaneously.
Leptin should be quantified in patients with KOA who do not meet the
diagnostic criteria for MS, to determine a group of KOA patients at high
risk of MS, for early intervention.
* New contributions of this doctoral thesis
This was the first scientific study in Vietnam on MS, plasma leptin
and IL-1β concentrations in patients with primary KOA. The prevalence

of MS in KOA was 51.7% and its increased when the stage of KOA
increased.
Determination of plasma leptin and IL-1β concentrations of the KOA
patients were higher than the control; the correlation between leptin and
IL-1β were determined with some metabolic risk factors. Determination
of the cut-off point of plasma leptin concentrations that predicts a patient
with knee osteoarthritis who may suffer from MS.
* The doctoral thesis arrangement: This thesis contains 125 pages
(without references and appendixes): Introduction: 02 pages, Chapter
1 Overview: 35 pages, Chapter 2 Subjects and methods: 28 pages,
Chapter 3 Results: 32 pages, Chapter 4 Discussion: 37 pages,
Conclusion: 02 pages, Recommendations: 01 page. It includes 33
tables, 9 graphs, 16 figures, 1 diagram and 174 references (7
Vietnamese references and 167 English references).


3
CHAPTER 1: OVERVIEW
1.1. Overview of osteoarthritis
1.1.1. Diagnosis of osteoarthritis
Diagnosis of OA was based on the ACR 1991 criteria.
1.1.2. New views of the pathogenesis of osteoarthritis
1.1.2.1. Low-stage chronic inflammation
Adipokines (leptin, resistin ...) and pro-inflammatory cytokines
(TNF-α, IL-1β, IL-6 ...) were related to the common mechanism of MS
and OA. Hypothesis on the independent role of low-stage chronic
systemic related to obesity and MS were caused the onset of OA.
1.1.2.2. Pro-inflammation cytokine (IL-1β)
IL-1β was a strong pro-inflammatory cytokine, important in the
early stages of OA because it was not only increased cartilage

catabolism (inhibiting type II collagen and aggrecan synthesis leads
to substrate degeneration) but also inhibited cartilage anabolic. IL-1β
stimulated IL-6 and IL-8 production; contributed to inflammation
(localized in synovial or systemic inflammation). Chondrocytes were
the main target cells of IL-1β, degenerative chondrocytes were higher
sensitive to the effects of IL-1β 3 to 4-fold than normal chondrocytes.
Only 1% of IL-1β receptors on active cartilage surface can convert
chondrocytes into a form of strong catabolism.
1.1.2.3. Adipokine (Leptin)
Leptin was an energy-regulating hormone, also a proinflammatory factor that caused inflammation and catabolism of
chondrocytes. Leptin resistance was the action of leptin that failed
when leptin concentrations increased. There was still controversy
about leptin causing degeneration or repair of cartilage.
Cartilage: Leptin acts on both cartilage anabolism and catabolism.
Leptin acts on chondrocytes through mechanisms of inflammation


4
and catabolism, producing IL-1β, MMP-9, MMP-13 ..., contributing to
cartilage degradation. Leptin caused cell proliferation, increased collagen
synthesis, stimulated endothelial calcium, increased bone mineralization,
increased growth factor IGF-1 and TGF-β1 (Dumond H.).
Bone: Leptin regulated bone growth indirectly through the neural
network, inhibiting bone formation. Leptin directly enhanced cells that
develop cortical bone, collagen synthesis and bone mineralization,
stimulating the formation of cartilage.
1.2. Metabolic syndrome
Metabolic syndrome, whose elements include central obesity,
dyslipidemia, impaired fasting glucose, and hypertension, increases
the risk of cardiovascular disease and mortality. In this study, MS was

diagnosed according to IDF criteria.
1.3. Relationship between metabolic syndrome and osteoarthritis
1.3.1. Obesity and osteoarthritis
Obesity contributed to OA through low-biomechanics and chronic
inflammation. Adipose tissue produced adipocytokines such as
leptin... in combination with pro-inflammatory cytokines such as IL-1
derived from macrophages in adipose tissue, causing inflammation of
synovial membranes, cartilage degeneration, changes in subchondral
bone that caused OA.
1.3.2. Glucose, insulin resistance and osteoarthritis
Chronic hyperglycemia caused oxidative stress, increased proinflammatory cytokine production, AGEs accumulation in joint
tissues, and differentiation of potential stem cells, insulin resistance
at local synovial membranes and low-stage chronic inflammation.
1.3.3. Lipid and osteoarthritis
There were many epidemiological and experimental evidence of


5
dyslipidemia related to OA. Lipid deposition in chondrocytes
occurred early. Another hypothesis was oxidized LDL (oxLDLs),
which caused atherosclerosis, promoted inflammation in OA.
Increasing free fatty acids can cause insulin resistance.
1.3.4. Hypertension and osteoarthritis
Hypertension caused endothelial dysfunction; reduction of
neural dynamics, localized perfusion reduction in peripheral
organizations including cartilage, reduction of oxygen and nutrient
supply, reduction of metabolism in cartilage. In addition, local
ischemia led to apoptosis of subcutaneous bone cells and subcortical
bone abnormalities.
1.4. Recent researches

1.4.1. Metabolic syndrome and osteoarthritis researches
According to the data of NHANES III, there were 7714 subjects,
the prevalence of MS was 59% in OA group compared with 23% in
the non-OA, one person with OA in middle age had a risk of MS
increased 5.26-fold.
1.4.2. Leptin and osteoarthritis researches
Leptin-deficient (ob/ob) and leptin receptor-deficient (db/db) female
mice compared with wild-type mice were studied. Extreme obesity due
to impaired leptin signaling induced alterations in subchondral bone
morphology without increasing the incidence of knee OA. Systemic
inflammatory cytokine levels remained largely unchanged in ob/ob and
db/db mice. These findings suggest that body fat, in and of itself, may
not be a risk factor for joint degeneration, because adiposity in the
absence of leptin signaling is insufficient to induce systemic
inflammation and knee OA in female mice (Griffin T.M.)
1.4.3. Leptin, metabolic syndrome and osteoarthritis researches
Increased leptin levels in both women and men with MS. Leptin


6
predicted that MS was independent of obesity. Leptin increased in
MS, correlated with the number of components of MS and leptin was
an important risk factor for KOA in women. Insulin resistance was
associated with increased KOA in men, increased leptin levels was
associated with increased KOA in women.
1.4.4. IL-1β and osteoarthritis researches
Ning L. et al found IL-1β concentrations related to the severity of
the disease so it was a marker for the severity of OA. Nguyen Ngoc
Chau found that plasma IL-1β concentrations in OA patients was higher
than the controls.

CHAPTER 2: SUBJECTS AND METHODS
2.1. Subjects
582 primary KOA patients and 78 healthy control individuals
were recruited during 2014 - 2019, at Bachmai hospital.
2.1.1. Study group
Inclusion criteria: Diagnosed primary KOA based on the ACR
1991 and MS based on the IDF 2005 criteria.
Exclusion criteria: secondary KOA, patients did not consent.
2.1.2. Controls
78 healthy individuals.
2.2. Research methods
2.2.1. Study design
Prospective, descriptive cross-sectional study.
2.2.2. Sampling method
Sample size for aim 1: p = 0.59 d = 0.04 ∝ = 0.05 n = 580.8 We
selected 582 patients in the KOA group
Z2(1-∝/2) × p × (1-p)
n=
d2
Sample size for aim 2: σ = 0.3 d = 0.05 ∝ = 0.05 n = 138.3
Z2(1-∝/2) × σ2
n=
d2
We selected 164 patients in the KOA* and 78 healthy individuals
in the controls. The KOA* was similar to KOA about age, BMI,
gender rate and the prevalence of MS.


7
2.2.3. Quantification of leptin and IL-1β plasma

Plasma leptin test by the Human leptin ELISA kit of Sigma;
plasma IL-1β test by Human IL-1β ELISA kit of Melsin, using
human monoclonal antibodies.
2.3. Data processing
Using medical statistics softwere: SPSS 20.0
2.4. Research diagram
Diagnosis of KOA

WC, blood pressure, HDL-C,
triglyceride, glucose, HbA1c, insulin
…

KOA*
(n = 164)

KOA
(n = 582)

CONTROLS
healthy people
(n = 78)
Leptin, IL1β

To determine the prevalence
of MS, its components and
the relationship with the
stages of primary KOA.

2. 2. To determine the relationship of
plasma leptin and IL-1β concentrations

with some clinical and subclinical features
in patients with primary KOA.

RECOMMENDATION
S


8
CHAPTER 3: RESULTS
3.1. Demographic, anthropometric, biochemical and
clinical characteristics in knee osteoarthritis group
Table 3.1. In 582 KOA patients, women were 86.6%; mean ages
were 56.7 ± 8.2 years; mean BMI was 24.0 ± 3.0 kg/m 2.
Table 3.6. No statistical differences between the mean ages, BMI,
WC, WHR, hypertension, HbA1c, CRP between men and women
groups.
3.2. The prevalence of metabolic syndrome and the relationship
with the stages of primary knee osteoarthritis
Table 3.8. The prevalence of MS and its components by sex
Total
(n = 582)
n (%)

Women
(n = 504)
n (%)

Men
(n = 78)
n (%)

MS

301 (51.7)

279 (55.4)

22 (28.2)

High WC

415 (71.3)

392 (77.8)

23 (29.5)

Low HDL-C

314 (54.0)

283 (56.2)

31 (39.7)

Hypertension

374 (64.3)

321 (63.7)

53 (67.9)

254 (43.6)

220 (43.7)

34 (43.6)

329 (56.5)

279 (55.4)

50 (64.1)

201 (34.5)

175 (34.7)

26 (33.3)

Variables

High fasting
glucose
High
triglycerides
Obese

p women - men

OR (95% CI)
< 0.001
3.2 (1.9 - 5.3)
< 0.001
8.4 (4.9 - 4.2)
< 0.05
1.9 (1.2 - 3.2)
> 0.05
0.83 (0.5 - 1.4)
> 0.05
1.0 (0.6 - 1.6)
> 0.05
0.7 (0.4 -1.1)
> 0.05
1.1 (0.6 -1.8)

The prevalence of MS among KOA was 51.7%, it was statistically
higher in women than men, with odds ratio was 3.2. The prevalence
of high WC, low HDL-C in women with KOA was significantly
higher than in men. No statistical differences of the prevalence of


9
hypertension, high fasting glucose, high triglycerides, obese between
men and women.

Chart 3.1. The prevalence of MS and its components by obese
The prevalence of MS, high WC, hypertension, high triglyceride,
low HDL-C in the obese was higher than in the non-obese, but
prevalence of hyperglycemia in two groups were the same.

10
Chart 3.4. The prevalence of metabolic syndrome increased when
the knee osteoarthritis stages increased.
Table 3.13. The prevalence of MS, high WC, hypertension, high
fasting glucose, high triglycerides in the late stage was statistically
higher than in the early stage.
Late
Early
(n = 148)
(n = 434)
Variables
p
OR (95% CI)
n (%)
n (%)
MS
95 (64.2) 206 (47.5) < 0.001 2.0 (1.4 - 2.9)
High WC
123 (83.1) 292 (67.3) < 0.001 2.4 (1.5 - 3.9)
Hypertension 111 (75.0) 263 (60.6) < 0.05 2.0 (1.3 - 3.0)
Hyperglycemia 76 (51.4) 178 (41.0) < 0.05 1.5 (1.04 - 2.2)
High TG
96 (64.9) 233 (53.7) < 0.05 1.6 (1.1 - 2.3)
Low HDL-C
83 (56.1) 231 (53.2) > 0.05 1.1 (0.8 - 1.6)
3.3. Plasma leptin and IL-1β concentrations
3.3.1. Characteristics of research groups
Table 3.15. The KOA* was similar to the KOA in mean age, BMI,

gender ratio and the prevalence of MS. The KOA* was similar to the
controls in gender ratio, but the difference in mean age, BMI and the
prevalence of MS.
3.3.2. Plasma leptin and IL-1β concentrations in primary knee
osteoarthritis compared with controls.
Table 3.16.:
Leptin, IL-1β, IL-1β/leptin in KOA* by sex
Variables (n = 164)
Leptin
(ng/mL)
IL-1β
(pg/mL)
Tỉ số
IL-1β/leptin

Median
Q1 - Q3
Median
Q1 - Q3
Median
Q1 - Q3

Total
(n = 164)
9.5
5.8 - 14.3
10.0
8.8 - 12.8
1.2
0.7 - 2.2

Men
(n = 23)
4.0
0.9 - 10.4
9.7
9.1 - 10.6
2.3
1.1 - 9.9

Women
(n = 141)
10.6
6.3 - 14.9
10.0
8.6 - 14.1
1.0
0.6 - 2.0

p
< 0.001
> 0.05
= 0.001

KOA*: Leptin concentrations in women were higher than in men;
IL-1β concentrations were not statistically different; IL-1β/leptin
ratio in men was higher than in women.


11

Table 3.17. Leptin, IL-1β, IL-1β/leptin in KOA* by obese
Variables

Obese (n = 59)

Non obese (n = 105)

(n = 164)

Median (Q1 - Q3)

Median (Q1 - Q3)

Leptin
(ng/mL)

13.0 (10.4 - 15.8)

7.9 (4.5 - 11.5)

< 0.001

IL-1β (pg/mL)

9.7 (8.8 - 12.1)

10.1 (8.6 - 13.4)

> 0.05

IL-1β /leptin

0.8 (0.6 - 1.5)

1.4 (0.9 - 3.3)

< 0.001

p

KOA*: Leptin concentrations in the obese were higher than in the
non-obese; IL-1β were not statistically different. IL-1β/leptin ratio in
the non-obese was higher than in the obese.
Table 3.18. Leptin, IL-1β, IL-1β/leptin in KOA* by MS
OA with MS (n = 85)

Variables

OA without MS (n = 79)

p

(n = 164)

Median (Q1 - Q3)

Median (Q1 - Q3)

Leptin (ng/mL)

11.6 (8.7 - 15.7)

7.7 (3.9 - 11.4)

< 0.001

IL-1β (pg/mL)

10.3 (8.9 - 13.5)

9.8 (8.6 - 12.0)

> 0.05

IL-1β/leptin

1.0 (0.6 - 1.7)

1.4 (0.8 - 3.4)

< 0.01

KOA*: Leptin concentrations in MS group were higher than the
non-MS; IL-1β were not statistically different. IL-1β/leptin ratio in
the KOA without MS was higher than the KOA with MS.
Table 3.21. Leptin, IL-1β, IL-1β/leptin in controls
Variables

Tổng

Men

Women

(n = 78)

(n = 11)

(n = 67)

Leptin

Median

0.5

0.1

0.5

(ng/mL)

Q1 - Q3

0.3 - 0.7

0.03 - 0.2

0.4 - 0.8

IL-1β

Median

6.9

6.6

6.9

(pg/mL)

Q1 - Q3

6.4 - 7.5

6.3 - 7.5

6.4 - 7.6

Tỉ số

Median

13.9

132.1

13.5

p
< 0.001
> 0.05
< 0.001


12
IL-1β /leptin

Q1 - Q3

9.6 - 24.9

27.2 - 227.5

9.4 - 19.2

Controls: Leptin concentrations in women was significantly higher
than in men; IL-1β were not significantly different; IL-1β/leptin ratio in
the men was higher than in the women.
Table 3.22. Leptin, IL-1β, IL-1β/leptin in KOA* and controls
KOA* (n = 164)

Controls (n = 78)

Median (Q1 - Q3)

Median (Q1 - Q3)

Leptin (ng/mL)

9.5 (5.8 - 14.3)

0.5 (0.3 - 0.7)

< 0.001

IL-1β (pg/mL)

10.0 (8.8 - 12.8)

6.9 (6.4 - 7.5)

< 0.001

IL-1β/leptin

1.2 (0.7 - 2.2)

13.9 (9.6 - 24.9)

< 0.001

Variables

p

KOA*: Plasma concentrations of leptin and IL-1β were higher,
IL-1β/leptin ratio was lower than in the healthy controls
Table 3.23. Leptin, IL-1β, IL-1β/leptin in three groups

Variables

OA with MS OA without MS

Controls

(n = 85)

(n = 79)

(n = 78)

Leptin

Median

11.6

7.7

0.5

(ng/mL)

(Q1 - Q3)

(8.7 - 15.7)

(3.9 - 11.4)

(0.3 - 0.7)

IL-1β

Median

10.3

9.8

6.9

(pg/mL)

(Q1 - Q3)

(8.9 - 13.5)

(8.6 - 12.0)

(6.4 - 7.5)

IL-1β
/leptin

Median

1.0

1.4

13.9

(Q1 - Q3)

(0.6 – 1.7)

(0.8 – 3.4)

(9.6 - 24.9)

p

< 0.001

< 0.001

< 0.001

The concentrations of leptin and IL-1β tended to decrease in three
groups: KOA with MS, KOA without MS and controls group (p <
0.001). In contrast, IL-1β/leptin ratio tended to increase.


13
3.3.3. Relationships between plasma leptin and IL-1β
concentrations with some clinical and subclinical features in
patients with primary knee osteoarthritis.
Table 3.24. Correlation between leptin with metabolic syndrome
components, obesity and some variables.

Leptin and
variables

Women

Men

KOA*

(n = 141)

(n = 23)

(n = 164)

r

p

r

p

r

p

Components of metabolic syndrome
SBP (mmHg)

0.134

> 0.05

0.030

> 0.05

0.036

> 0.05

DBP (mmHg)

0.091

> 0.05

0.004

> 0.05

-0.019

> 0.05

Glucose (mmol/L)

0.056

> 0.05

0.163

> 0.05

-0.013

> 0.05

Triglyceride (mmol/L)

0.080

> 0.05

0.047

> 0.05

0.014

> 0.05

HDL-C (mmol/L)

0.054

> 0.05

-0.217

> 0.05

0.087

> 0.05

WC (cm)

0.430

< 0.001

0.723

< 0.001

0.417

< 0.001

Weight (kg)

0.459

< 0.001

0.657

= 0.001

0.381

< 0.001

BMI (kg/m2)

0.514

< 0.001

0.548

< 0.01

0.489

< 0.001

Insulin (µU/mL)

0.420

< 0.001

0.668

< 0.001

0.403

< 0.001

HbA1c (%)

0.303

< 0.001

0.114

> 0.05

0.185

< 0.05

HOMA-IR

0.383

< 0.001

0.626

= 0.001

0.346

< 0.001

IL-1β (pg/mL)

-0.046

> 0.05

-0.204

> 0.05

-0.046

> 0.05

CRP (mg/L)

0.113

> 0.05

-0.051

> 0.05

0.037

> 0.05

Age (năm)

0.113

> 0.05

-0.337

> 0.05

0.028

> 0.05

Leptin concentrations were positively correlated with markers of
obesity and insulin resistance (WC, weight, BMI, insulin, HOMA-IR);
were not correlated with MS components (except WC), IL-1β, CRP, age.


14
Table 3.25. Multivariate regression equation between leptin with
independent variables in women group
Variables
Beta coefficient standardized
p
Insulin (µU/mL)
0.760
< 0.001
BMI (kg/m2)
0.216

< 0.01
HOMA-IR
-0.397
< 0.05
R2 = 0.298 và p ANOVA < 0.001
Leptin = 1.462*insulin + 0.638*BMI – 2.402*HOMA-IR -11.209
Women: insulin, BMI, HOMA-IR affected about 29.8% of the
plasma leptin concentrations, in which insulin had the strongest effect.
Table 3.26. Correlation between IL-1β, IL-1β/leptin with CRP and
some other variables.
IL-1β
Tỉ số IL-1β/leptin
Variables
(n = 164)
r
p
r
p
CRP (mg/L)
-0.100
> 0.05
-0.009
> 0.05
VAS
0.086
> 0.05
0.204
< 0.01
Age (năm)
0.101

> 0.05
-0.012
> 0.05
Components of metabolic syndrome
WC (cm)
-0.086
> 0.05
-0.365
< 0.001
SBP (mmHg)
-0.002
> 0.05
-0.057
> 0.05
DBP (mmHg)
-0.064
> 0.05
-0.028
> 0.05
Glucose (mmol/L)
-0.070
> 0.05
-0.047
> 0.05
Triglyceride (mmol/L)
0.153
= 0.05
0.066
> 0.05
HDL-C (mmol/L)

-0.042
> 0.05
-0.118
> 0.05
BMI (kg/m2)
-0.029
> 0.05
-0.390
< 0.001
HbA1c (%)
-0.110
> 0.05
-0.218
< 0.01
Insulin (µU/mL)
-0.180
< 0.05
-0.432
< 0.001
HOMA-IR
-0.189
< 0.05
-0.388
< 0.001
IL-1β concentrations were not correlated with inflamatory marker
(CRP), VAS, all components of MS, all markers of obesity and age;
Weak negatively correlation with insulin, HOMA-IR. IL-1β/leptin
ratio inversely correlatedwith WC, BMI, Insulin, HbA1c, HOMA-IR
(r from -0,432 to -0,365; p < 0.01).

15
Table 3.27.
Variables
(n = 164)
1
2
3
4

4 stages of Xray

Early
Late

Early and late

0
1
2
3
4
5

Number of MS
components

Leptin (ng/mL)
Median (Q1 - Q3)

IL-1β (pg/mL)
Median (Q1 - Q3)

9 (4.2 - 13.1)
9.6 (5.8 - 14.8)
9.9 (5.7 - 14.3)
20.8 (9.4 - 40.5)
p > 0.05
9.4 (5.7 - 14.0)
10.4 (6.0 - 14.6)
p > 0.05
7.6 (1.2 - 11.1)
7.7 (1.7 - 12.4)
8.3 (4.2 - 11.4)
11.1 (6.0 - 15.0)
9.5 (6.1 - 15.7)
11.5 (7.5 - 16.2)
p > 0.05

9.6 (8.6 - 14)
10.1 (8.7 - 13.6)
10.4 (9.1 - 12.5)
9.6 (9 - 11.4)
p > 0.05
9.9 (8.7 – 13.8)
10.1 (9.1 - 12.2)
p > 0.05
9.8 (8.3 - 10.4)
10 (8.5 - 17.9)
9.7 (8.4 - 11.8)

10.3 (8.4 - 14.3)
10.0 (9.1 - 12.4)
10.0 (9.0 - 15.3)
p > 0.05

Median of leptin or IL-1β concentrations were not statistically
different in the groups of Xray, early and late, number of MS components.
Table 3.28.
Variables
(n = 164)
BMI
(kg/m2)

Number of
KOA
Duration

Leptin (ng/mL)
Median (Q1 - Q3)

IL-1β (pg/mL)
Median (Q1 - Q3)

< 23

6.2 (3.4 - 10.3)

10.2 (8.5 - 13.8)

23 - 24.9

9.4 (6.9 - 12.9)

10.1 (8.7 - 13)

≥ 25

13 (10.4 - 15.8)
p < 0.001
8.5 (4.0 - 11.9)
10.5 (6.3 - 15.7)
p < 0.05
9.2 (5.4 - 13.3)
11.9 (8.7 - 15.6)
p < 0.05

9.7 (8.8 - 12.1)
p > 0.05
9.6 (8.2 - 12.2)
10.3 (9 - 13.2)
p > 0.05
9.2 (8.7 - 14.0)
10 (8.9 - 11.0)
p > 0.05

1 knee
2 knees
Short
Long

16
Median of leptin concentrations increased gradually when BMI
increased, number of KOA, the duration of KOA was long; median of
IL-1β concentrations were not statistically different in these groups.
Table 3.31. The cut-off point of plasma leptin to predict MS.
Leptin

Cut-off

(ng/mL)

point

Men
n = 23
women
n =141

AUC

95% CI

p

Sensitivity

Specificity

5.5

0.892

0.744 - 1.0

< 0.01

1.0

0.867

8.7

0.643 0.55 - 0.736 < 0.01

0.779

0.516

The cut-off point of plasma leptin concentrations in men KOA
was 5,5 ng/mL, in women KOA was 8.7 ng/mL to predict MS.
CHAPTER 4: DISCUSSION
4.1. General characteristics of patients with
knee osteoarthritis
4.1.1. Anthropometric characteristics
The mean age was 56.7 ± 8.2, from 37 to 81 years old. 86.6%
were women. Many researchs found that OA was more common in
women than men, especially after menopause, which may be related
to estrogen deficiency or due to imbalance between bone formation
and bone destruction related to leptin.

Mean BMI was 24.0 ± 3.0 kg/m 2, from 16.3 to 35.7 kg/m2; The
proportion of overweight and obese accounted for 61.5%, similar to
Bui Hai Binh's 60.7%. Mean BMI of men was similar to women
group. Many researchs find the role of weight or BMI affecting the
formation and development of OA, especially in the knee joint.


17
4.1.2.Clinical and subclinical symptoms
582 KOA patients with 905 ostoarthritis of the knee joints, the
right knee was 446 (49.3%), equivalent to the number of the left knee
was 459 (50.7%). All patients had joint pain and short morning
stiffness for less than 30 minutes or no stiffness. Symptoms of red,
hot, swelling, muscular atrophy had below 20%. In addition, valgus
was more common than varus deformity, the mean VAS and
WOMAC scores were lower than the intervention studies. The mean
plasma CRP concentrations was 2.9 ± 5.3 mg/L, similar to SanchezRamirez D.C. was 2.9 ± 3.1 mg/L. The mean CRP in the MS group
was higher than the non-MS. The rate of OA in the early stage (1 and
2) was 75.5%.
4.2. The prevalence of metabolic syndrome and the relationship
with the stages of primary knee osteoarthritis
4.2.1.Prevalence of metabolic syndrome in knee osteoarthritis
MS was present in 51.7% KOA patients using the IDF criteria. This
prevalence was similar to study of Shin D. in Korea (52.4%), ElSaid
T.O. et al in Egypt (53.7%), Abourazzak F. et al in Morocco (48.5%);
was higher than study of Xie D.X. et al in China (20.3%); was lower
than study of Yerima A. et al in Nigeria (59.8%) and Puenpatom R.A. et
al in the US (59%). In many researchs, the ratio of MS in OA was
change from 20% to 60%, because the researches used different MS
criteria, different OA criteria such as based on clinical and Xray or only

based on Xray, different OA stages ...
In women, the incidence of MS in KOA was 55.4%, it was higher
than men (28.2%), an odds ratio of 3.2 and 95% CI of 1.9 - 5.3 (p <
0.001), similar to ElSaid T.O et al study (62.4% in women, higher
than mens), differ from Niu J., et al the prevalence of MS in men OA
was 26.7% higher than in women (22.9%) and Maddah S. et al, the
prevalence of MS in mens (24.0%) was higher than women (18.3%).
Firstly, due to the application of different criteria of MS, we used IDF
to take WC as a compulsory standard, while other studies used NCEP
to assess equally 5 criteria without mandatory criteria. Secondly, our


18
women patients had a mean WC higher than the 80 cm, while the mean
WC in the men group was lower than the 90 cm, threshold, the
prevalence of MS in the women OA was much higher than men. The
incidence of MS in KOA with obesity was 72.6%, higher than nonobesity KOA (40.7%) with an odds ratio of 3.9. IDF considers obesity to
be the main cause of MS. Obesity contributes to hyperglycemia,
hypertriglyceridemia, insulin resistance, low HDL-C, hypertension.
The incidence of MS increases when the OA stage was getting worse
(Chart 3.4), which means that when OA was getting worse, the patient
had more comorbid conditions such as hypertension, impaired fasting
glucose, dyslipidemia. The incidence of MS in severe KOA was
significantly higher than mild with OR by 2.0 and 95% CI from 1.4 to
2.9 (p < 0.001) (Table 3.13), similaly to Vasilic-Brasnjevic S. et al.
4.2.2. Relationship between obesity and knee osteoarthritis
Obesity was the biggest risk factor for KOA and was an important
component of MS. The incidence of MS in obesity was higher than
the non-obesity group (72.6% vs 40.7%). In this study, the obesity
rate was 34.5%; BMI was 24.0 ± 3.0 kg/m 2; increased WC rate was

71.3%; the prevalence of MS in women group was higher than men
because of the rate of increased WC in the women was higher than
the men group. The incidence of both sides KOA in the obese was
significantly higher than the one side group with odds ratio of 1.5;
Similar conclusions of Sellam J. obesity increased the risk of KOA,
especially OA knees on both sides. We also found abdominal obesity
associated with OA but not BMI; Suitable for Vasilic-Brasnjevic et
al, abdominal obesity or BMI ≥ 30 kg/m2 strongly correlated with the
Xray stage. Obesity increased the chronic mechanical pressure at
load bearing joints. Adipocytokine production, adipocytokines such
as leptin, adiponectin ... they combined with pro-inflammatory
cytokines such as IL-1β, TNF-α ... derived from adipose tissue,
causing inflammation of synovial membranes, cartilage degradation;
leptin acted directly on connective tissue to promote formation and
progression of OA.
4.2.3. Relationship between hypertension and knee osteoarthritis


19
The concept of "hypertension" according to the IDF criteria for
MS, very differed from the JNC 8. According to IDF, the
hypertension rate of KOA was lower than Puenpatom R.A. et al
(64.3% vs 77.7%). There were very few studies that explain the
relationship between hypertension and OA. A hypothesis
hypertension caused endothelial dysfunction; reduction of neural
dynamics, localized perfusion reduction in peripheral organizations
including cartilage, reduction of oxygen and nutrient supply,
reduction of metabolism between cartilage and covered cartilage. In
addition, ischemia leaded to apoptosis of subchondral cells and
turnover abnormalities.

4.2.4. Relationship between plasma sugar and knee osteoarthritis
Glucose: The concept of "hyperglycemia" according to the IDF
criteria for MS was different from IDF criteria for diabetes. The rate
of hyperglycemia of patients with KOA was higher than Puenpatom
R.A. study (43.6% vs 30.7%). It may be due to differences in size
and age, glycemia concentrations increased with age.
• Insulin concentrations: The insulin concentrations of the MS
group were higher than the non-MS.
• HOMA-IR: mean HOMA-IR in the MS group was higher than
the non-MS, the men was lower than the women group. According to
the two meta-analyzes of Louati and Williams, type 2 diabetes was a
risk factor for OA. Of all components of MS, diabetes alone was
considered an independent risk factor for progression of KOA. There
were two types of type 2 diabetes mechanism acting on joint tissue.
Firstly, chronic hyperglycemia causes oxidative stress, increased proinflammatory cytokine production, accumulation of AGEs in joint
tissue, and differentiation of potential stem cells. Secondly, insulin
resistance in local synovial membrane of diabetic patients and lowstage systemic inflammation associated with insulin resistance.
4.2.5. Relationship between dyslipidemia and knee osteoarthritis


20
Degenerative cartilage reduced regulating gene expression,
absorption of cholesterol into cells causes accumulation of cholesterol in
chondrocytes (Tsezou A. et al). Increasing free fatty acids can cause
insulin resistance. Treatment of dyslipidemia with statins in OA patients
for 2 years that reduce in progress of OA (Wang Y).
• Triglyceride: The prevalence of triglyceride in KOA patients
was 56.5%, there were no difference between men and women. The
rate of hypertriglyceride in the obese was statistically higher than the
non-obese group; in the group with MS was statistically higher than

the non-MS, in the late was statistically higher than early stage.
• Low HDL-C: The rate of low HDL-C in the MS group was
statistically higher than without MS group. There was no difference
in rate of low HDL-C between late and early groups.
The rate of increased triglyceride (56.5%) was equivalent to the
low HDL-C (54%), similar to Bui Hai Binh study (52.4%).
Compared to Gweressus E.D. et al, our mean triglyceride was higher,
the mean HDL-C was lower. This difference may be due to race,
Vietnamese people tend to eat more starch, so they tend to have a
higher rate of increased triglycerides and lower HDL-C.
4.3.Plasma leptin and IL-1β concentrations
4.3.1.Characteristics of disease groups, disease groups*, controls
KOA group and KOA* group were similar to gender, age, BMI
and rate of MS indicate that the KOA* group was representing the
KOA group. KOA* group and controls were similar to rate of sex but
differences in BMI and rate of MS
4.3.2. Plasma leptin
4.3.2.1. Leptin concentrations in patients with primary knee
osteoarthritis
KOA* group: leptin concentrations was between 0.07 - 75.8
ng/mL, median 9.5 ng/mL; interval quartile 5.8 - 14.3 ng/mL; mean
11.5 ± 10.0 ng/mL; it was similar to result of Zheng S. (10.65 ± 13.10
ng/mL); lower than most published results, probably due to


21
differences in obesity criteria, mean BMI in these studies were higher
in our study, which leptin correlated with BMI.
Leptin concentrations in KOA* group had median value and interquartile range of 9.5 (5.8 - 14.3) ng/mL, significantly higher than the
controls of 0.5 (0.3 - 0.7) ng/mL. Leptin concentrations were highest

in the KOA group with MS, higher than the KOA group without MS,
the lowest in the healthy controls (p < 0.001). This suggests that
leptin had an intermediate role in the relationship between leptin and
MS and OA. However, this was a cross-sectional study, so it was not
possible to determine the causal relationship of leptin-induced OA or
OA causing increased leptin-associated MS.
4.3.2.2. Factors affecting leptin concentrations in KOA patients
• Leptin and sex: In women, leptin concentrations were higher
than men in both KOA* and controls groups. In women, leptin was
always higher than men, regardless of OA status. Due to the effects
of sex hormones, leptin was two to three-fold higher in women than
men, who had the same amount of fat.
• The correlation between leptin and BMI: a moderate positive
correlation between leptin and BMI (r = 0.489 and p < 0.001), similar
to result of Staikos C. et al (r = 0.49 and p < 0.01) and slightly less
than Manoy’ results (r = 0.57 p < 0.001).
• Correlation between leptin and WC: moderate positive
correlation between leptin and WC (r = 0.417 and p < 0.001), similar
to result of Manoy P. (r = 0.49 and p < 0.001), this correlation was
even more powerful if considered separately in the men group, the
correlation was tight (r = 0.723 and p < 0.001).
• Correlation between leptin and weight: moderate positive
correlation between leptin and weight (r = 0.459 and p < 0.001);
similar to result of Iwamoto J (r = 0.392 and p = 0.0084).
• Leptin and insulin resistance: a positive correlation between
leptin and HOMA-IR (r = 0.346 and p < 0.001); particularly, in men
group achieved a relatively positive correlation (r = 0.626 and p =
0.001), moderate positive correlation in the women; A moderate

22
positive correlation between leptin and insulin (r = 0.403 and p <
0.001); particularly, the men group correlated quite tightly (r = 0.668
and p = 0.001), in the women, the correlation was medium. There
was no correlation between leptin and glucose and HbA1c, similar to
two studies of Vo Minh Phuong and Nguyen Van Hoan.
• Leptin and blood pressure, lipids profile: no correlation.
4.3.2.3. The cut-off points of leptin predict the risk of MS
In women KOA group: the cut-off point of leptin concentrations
that predicted MS was 8.7 ng/mL with AUC was 0.643; sensitivity
was 77.9%; specificity was 51.6% and p <0.01. Men KOA: the cutoff point of leptin concentrations that predicted MS was 5.5 ng/mL
with AUC was 0.892; sensitivity was 100%; specificity was 86.7%
and p <0.01. It was possible to use plasma leptin concentrations to
predict a patient with OA who was at high risk of suferring from MS
when patient was not eligible for IDF criteria.
4.3.3. IL-1β concentrations in primary knee osteoarthritis
4.3.3.1. IL-1β concentrations
KOA* group: IL-1β concentrations had range from 5.2 to 106.9
pg/mL, median 10 pg/mL; interval of quartile 8.8 - 12.8 pg/mL; mean
14.6 ± 15.9 pg/mL. In the controls, IL-1β had median 6.9 pg/mL;
interval quartile 6.4 - 7.5 pg/mL. Thus, IL-1β concentrations in the
KOA* group was higher than the controls (p < 0.001), this result was
similar to Nguyen Ngoc Chau’ result. The concentrations of IL-1β
tended to decrease in three groups: OA with MS, OA without MS and
controls (p < 0.001). Thus, accumulating more health problems such
as OA and MS. Many studies repored that IL-1β concentrations
changed in a very wide range, a hundres of times, our IL-1β results
were in this range.
4.3.3.2. Relationship between IL-1β and leptin
No correlation, this differ from Hussein N.A. There was a strong

correlation between leptin and IL-1β (r = 0.85 and p < 0.0001) in 84
patients with obesity and primary KOA.
4.3.3.3. IL-1β/leptin ratio


23
IL-1β/leptin ratio had renge from 0.12 to 255.6, median 1.2;
interval quartile 0.7 - 2.2; mean 6.9 ± 28.7. IL-1β/leptin ratio in KOA*
group was lower than in the controls, it tend to increase from the OA
group with MS to the OA group without MS and controls. IL-1β/leptin
correlated negatively with WC, BMI, insulin, HOMA-IR (p < 0.001). In
fact, this ratio was less valuable.
4.3.3.4. Relationship between IL-1β concentrations and some
items:
IL-1β had almost no correlation or weak correlation with age,
anthropometric, blood pressure, lipid profiles, glucose, insulin
resistance, uric acid, CRP, WOMAC.
CONCLUSIONS
1. The
prevalence
of
metabolic
syndrome
(according to IDF), its components and the
relationship with the stages of knee osteoarthritis
- The prevalence of MS in KOA was 51.7%, its components such as
increased waist circumference, hypertension, high triglyceride, low
HDL-C, and hyperglycemia was 71.3%, 64,3%, 56.5%, 54.0%, and
43.6%, respectively. The prevalence of MS among women was higher
than men, in obese was higher than non-obese group (p < 0.001).

- The prevalence of MS in stage 1, 2, 3, and 4 were 34.7%, 55.4%,
63.5%, and 72.7%, respectively, it increased when the OA stages
increased (p < 0.001). The prevalence of MS, increased waist
circumference, hypertension, hyperglycemia, high triglyceride in the
late were higher in the early OA stage (p < 0.05).
2. Plasma leptin and IL-1β concentrations in primary KOA
2.1. Plasma leptin concentrations in knee osteoarthritis
- Leptin concentrations in KOA were 9.5 ng/mL; higher than in
controls was 0.5 ng/mL (p < 0.001). Leptin concentrations in women
were higher than in men group, in obese was higher than in nonobese group and in MS group was higher than in non-MS. Leptin