MINISTRY OF EDUCATION AND TRAINING

MINISTRY OF HEALTH

HANOI MEDICAL UNIVERSITY

NGUYEN TRONG KHAI

STUDY ON EPIDEMIOLOGICAL AND CLINICAL
CHARACTERISTICS OF DIABETIC
RETINOPATHY AND EFFECTIVENESS OF
INTERVENTION MEASURES IN HA NAM
PROVINCE

Field of study : Ophthalmology
Code

: 62720157

SUMMARY OF MEDICAL DOCTORAL THESIS

HANOI – 2018


THE THESIS WAS COMPLETED AT:
HANOI MEDICAL UNIVERSITY

Scientific advisors:
1. Assoc.Prof. Dr. Hoang Nang Trong
2. Assoc. Prof. Dr. Hoang Thi Phuc
Reviewer 1:

fastest growth. Diabetes causes many dangerous complications: including
acute complications and chronic complications. Common chronic
complications are cardiovascular diseases, eye diseases, kidney diseases
and neurological diseases etc.
Diabetic retinopathy (DR) is the most common complication of
diabetic eye disease. According to WHO, the incidence of DRs ranges
from 20% to 40% of people with diabetes, which varies by countries
and regions. Diabetes mellitus and blood glucose control are the major
risk factors of DR. This is the leading cause of vision loss and
blindness. People with diabetes have a 30% increased risk of blindness
compared to those at the same age and sex.
Currently, in Vietnam there have been some studies on diabetes,
DR, and factors related to these diseases in Vietnam. Besides, there
have been studies which refer and introduce modern and effective
treatment methods. However, community-based intervention programs
for prevention of diabetic complications are still limited, especially with
DR, the corresponding intervention programs and effectiveness
evaluations are almost inadequate. We therefore conducted the study
"Epidemiological and clinical characteristics of diabetic retinopathy
and effectiveness of intervention measures in Ha Nam province" with
two objectives:
1. Describe the epidemiological, clinical characteristics and some
factors related to diabetic retinopathy in diabetic patients being
managed in Ha Nam province in 2013.
2. Evaluate the effectiveness of intervention against diabetic
retinopathy in Binh Luc district, Ha Nam province.
2. New contributions of the thesis
The study results described epidemiological and clinical
characteristics of diabetic retinopathy in diabetic patients being managed in
Ha Nam province, a province in the Red River delta, where the people are

and methods of study (11 pages), Chapter 3: Study results (31 pages),
Chapter 4: Discussion (29 pages), Conclusion (2 pages),
Recommendations (1 page). Others: reference section, 7 appendices,
tables, charts and illustrations.
Chapter 1: LITTERATURE REVIEW
1.1. Epidemiological characteristics of diabetes
Diabetes is a syndrome characterized by hyperglycemia due to the
effect of lack of or complete loss of insulin or because of its association
with impairment in the secretion and activity of insulin.


3
There are many ways to classify diabetes, but the new
classification of WHO, based on disease type, is currently widely
applied: type-1 diabetes (about 5-10%), and type-2 diabetes (about 9095%). The complications of diabetes are usually divided by the time of
occurrence and the extent of complications: including acute and chronic
complications. Diabetic retinopathy (DR) is one of the common eye
complications
1.2. Epidemiological, clinical characteristics of diabetic retinopathy
1.2.1. Epidemiological characteristics of diabetic retinopathy
The DR disease develops in nearly all people with type-1 diabetes and
over 77% of people with type-2 diabetes over 20 years. Sobha (2012)
conducted cross-sectional studies in patients who examine diabetes in
hospitals. Among people with type-2 diabetes, the incidence of DR was
38% in the white European group, 52.4% in the African group, 42.3% in
the South Asian group. A study in Taiwan reported that new incidence in
the first year was 1.1% for women and 1.5% for men. In Viet Nam,
Nguyen Thi Thu Thuy (2009) conducted a study which shows that the
complication incidence of DR accounted for 28.7%.
1.2.2. Pathogenesis of diabetic retinopathy

vein to enter the glass chamber.
Today, there are many ways to classify the DR disease, but the
simple classification used the most is the Alphediam classification that
divides the disease into two main categories: non-proliferative DR and
proliferative DR. Venous jaundice can be found in both proliferative
and non-proliferative forms.
Retinopathy of diabetic retinopathy is characterized by mild, nonproliferating stages (at least one aneurysm and hemorrhage, no other
retinal lesions); Moderate hyperplasia (other extra lesions such as soft
tissue discharge, venous injury and microvascular abnormalities in the
retina); Severe proliferations (one or more of the following signs:
haemorrhage and multiple aneurysm over 4 quarters), venous
abnormalities seen in both quadrants, abnormalities of deep veins in the
retina encountered at least one quarter corner); and severe proliferation
(there are two signs of severe DR but no precursors). Proliferative
diabetic retinopathy is characterized by the following stages: early
proliferation (pre-necrotic neoplasia of less than half the area of the
visceral disk), high-risk proliferation with 3 moderate, evidence.
Venereal disease: It can be seen at every stage of the disease. The
crown is thickened, with diameters up to 2 times the disk diameter.
There are signs: follicular papillae, anemia of royal anemia.
1.2.4. Some factors related to diabetic retinopathy
The duration of diabetes is a leading risk factor for retinal
complications. In most cases, DR disease develops over 10-15 years. In
Vietnam, a study of Nguyen Thi Lan Anh (2017) showed that patients
with history of diabetes over 10 years had a 15.9 times higher risk of
developing the disease than those with diabetes less than 10 years.


5
The relationship between the quality of blood glucose control and

1.3. Interventions to prevent and treat diabetic retinopathy
Interventions to prevent and treat DR include two groups: direct
therapy and prevention / preventive programs


6
Treatment method of non-proliferative DR is mainly based on
optimizing the health of patients. The best treatments available today
are to prevent the progression and proliferation of the disease with blood
glucose control. Treatment of DR must have a close association between
the specialist eye, endocrine and cardiovascular medicine. Depending on
the lesion of the disease to have appropriate treatment. With advances in
the treatment of diseases of the glass, the retina in general, DR in
particularly. In the treatment of peripheral retinoblastoma treated with
retinal laser or glass, nowadays some new treatments such as corticoid or
vascular endothelial growth factor
In addition to the treatment interventions mentioned above,
currently, in the world as well as in Vietnam, there are not many
intervention studies which apply measures dedicated to the prevention
of DR. However, the literature review showed that controlling diabetes
is a way to prevent the complications of diabetes in general, and the DR
disease in particular in an effective manner. Controlling diabetes by
taking the right medicine, doing physical activity, and maintaining a
healthy diet can prevent or delay vision loss. Because the DR disease is
often unnoticed, the disease is only detected when the eyesight is
reduced, so people with diabetes should have a comprehensive eye
exam at least once a year. Early detection, timely treatment, proper care
and monitoring of diabetes can protect against vision loss.
In practice, the intervention programs are selected by the target
population, so that intervention programs may be biased towards

2016 in the Binh Luc district (intervention), Ly Nhan district
(comparison) of Ha Nam province.
2.1.1. Study objects of the objective 1
- Selection criteria: Patients diagnosed with type -2 diabetes who are
being managed in Ha Nam province.
- Exclusion criteria: Patients who did not agree to participate in the
study, patients whose eye bottom could not be clearly seen due to corneal
scars, cataracts etc.
2.1.2. Study objects of the objective 2
- Selection criteria: People with diabetes who do not have DR in
the group of diabetic patients, are examined in the stage 1.
- Exclusion criteria: Patients who did not agree to participate in the
study, patients not permanently residing in the study area, patients who
withdrew during follow-up in the community.


8
2.2. Methods of study
2.2.1. Study design
- Design of study of the objective 1: Cross-sectional study
- Design of study of the objective 2: intervention study with prepost comparison and comparison groups.
2.2.2. Sample size and sample selection
- Sample size for objective 1:
Sample size is determined by the formula:

Of which:
n: Study sample size
Z(1-α/2) : means the confidence level obtained at the probability level
α = 5% (equals 1.96).
P: means the average rate of diabetic retinopathy according to a

patients with an average incidence rate of 1.49% per year).
With assumption for withdrew objects during the study, we add 10%
of patients. Therefore, the minimum sample size for the study of
objective 2 will be 77 people per group. After the intervention, in 210
objects who were assessed at initial point of time (M1) and included in
objective 2 of the study, 13 objects were excluded from the study because
they did not regularly reside in the locality (6 objects), could not contact
(4 objects), did not agree to continue to participate in the study (3
objects). Therefore, only 197 objects were included as data for evaluating
the effectiveness of the intervention (the follow-up rate of the study was
93.8%). This sample size satisfies the quantitative requirement according
to the calculation of the sample size formula set for objective 2.
2.3. Equipment for study
Equipment for study includes the Landolt eyeglass, the test box,
the Goldmann manometer, examination microscope, direct, indirect
ophthalmoscope, Volk + 20D, + 90D glass, non-dilated pupil cataract,
A-B ultrasound, Mydrin-P 1% pupil dilution medicament, sample
medical records and interview forms.
2.4. Steps of study
Step 1: Information collection: via interview form, study disease
sample is conducted by Ha Nam Ophthalmology Hospital, Ha Nam
Health Communication and Education Center.
Step 2: Eye examination: Performed by Ha Nam Ophthalmology
Hospital staffs. There are 2 doctors in charge of measuring and
photographing the optic. The ophthalmoscopy was determined based on the
intraocular photograph and clinical examination. Cases of clinical


10
manifestations that are unclear need fluorescence or OCT sent to the Central

Quantitative information was collected from the patients using the
interview questionnaire, which was conducted by the staff of the Hanam
Ophthalmology Hospital and Ha Nam Health Education


11
Communication Center. Clinical information was collected after using
equipment assisted by Ha Nam Ophthalmology Hospital staff.
2.7. Processing data
- Data entry using Epi data 3.1, cleaning, data coding: using SPSS 22.0
software
- Using the χ2 test with large samples, accurate Fisher test with
small samples and statistical algorithms to find the correlation.
- Using paired sample t-test to compare the status of the same
objects before and after the intervention. Pairing creates constraints on a
number of anthropometric, epidemiological, and clinical factors. The
significant level of 0.05 were applied for all statistical analyzes.
- In order to evaluate the effectiveness of community-based
intervention programs, the efficiency indicator = [| P1-P2 | / P1] x 100% is
usually used. Intervention effectiveness index is calculated by the effect of
the two efficiency indicators in Binh Luc and Ly Nhan districts.
- The method of using generalized estimating equations (GEE) is
used to explore the overall effect of interventions on the change of DR
incidence rate adjusting with other risk factors.
2.8. Ethics in study
This thesis was approved by the Scientific Council of Hanoi Medical
University. With the consent of the Central Eye Hospital, the Provincial
People's Committee and the Health Department of Ha Nam Province. The
patients voluntarily participated in the study and their information is kept
confidential. Patients determined with eye injury should be treated, making



13

Figure 3.3: Percentage distribution of retinal damage
Patients with DR are 30.9%. Of these, 77.7% did not proliferate
mildly, 15.6% did not proliferate moderately and 5.9% did not
proliferate severely.
3.1.3. Characteristics of diabetes history
The rate of patients with regular check-up was 79.2%, strict
treatment 75.1% and 59.3% of patients with good diabetes treatment.
Most patients have a diabetes prolonged less than 5 years (42.6%).
Illness duration of 5-10 years and over 10 years had lower rates of
29.1% and 28.3%, respectively.
3.1.4. Para-clinical characteristics of the study objects
- Blood glucose level:

Figure 3.4: Distribution of blood glucose level


14
50.2% of patients had the normal blood glucose level, 28.6% had
blood glucose level of 7-9 mmol/l and 21.2% had blood glucose level
above 9 mmol/l.
Blood lipid, BMI indicator and glaucoma: Most patients have the
normal blood lipid level (91.5%). 35.6% of patients were overweight,
obese, and 5.9% were malnourished. The rate of hypertensive patients
was 49.2%
3.1.5. Some factors related to diabetic retinopathy.
- Relations of anthropometric and socioeconomic factors to

15
3.2.2. Incidence of diabetic retinopathy
Table 3.1: Distribution of incidence rate of diabetic retinopathy
before and after intervention
Binh Luc
M12
M24 Cumulative
n (%) n (%)
n (%)
No
new 100
97
97 (93.3)
patients
(96.2) (97.0)
New
4
3
7
patients
(3.8)
(3.0)
(6.7)
104
100
104
Total
(100)
(100)
(100)

eye. The sample size for comparative testing in Binh Luc was 208 eyes and
Ly Nhan 186 ones. Generally, there was no significant change in the
eyesight status of study objects before and after intervention (p> 0.05).
3.2.4. Changes in BMI, blood glucose and blood pressure
The mean blood glucose indicator in the intervention patients was
significantly reduced after intervention (p = 0.05).
Table 3.2: Test for change in BMI and blood glucose
before and after intervention

BMI (kg/m2)
pBMI
Blood glucose
(mmol/l)
Pblood glucose

Binh Luc
M1
M24
Mean (SD) Mean (SD)
22.0
22.1
(2.8)
(2.7)
0.77
7.8
6.7
(6.3)
(1.5)
0.05


71
50
No hypertension
60 (64,5)
(52.9)
(68.3)
(53,8)
49
33
43
Hypertension
33 (35,5)
(47.1)
(31.7)
(46,2)
104
104
93
93
Total
(100)
(100)
(100)
(100)
0.52
0.63
OR (95% CI)
(0.28 – 0.95)
(0.34-1.20)
IEhypertension

Treatment
regimen
Strict
69 (66.4) 94 (90,4) 61 (65,6) 68 (73,1)
OR (95%CI)
4.77 (2.12-11.47)
1.42 (0.72-2.81)
IEtreatment regimen
24.7%
Not good
41 (49.0) 27 (36,0) 37 (39,8) 41 (44,0)
Treatment
effectiveness Good
53 (51.0) 77 (74,0) 56 (60,2) 52 (56,0)
OR (95%CI)
2.21 (1.16-4.20)
0.8 (0.4-1.6)
IEtreatment effectiveness
38.1%
There was a positive change in the monitoring regimen, treatment
regimen and treatment effectiveness of patients with diabetes in the
intervention group (p
In Binh Luc, the better knowledge rate was 25%, the unchanged
knowledge rate was 64%, the poorer knowledge rate was 11%. The
better practice rate was 61%, the unchanged knowledge rate was 34%,
the poorer practice rate was 5%.
In Ly Nhan, the better knowledge rate was 8.3%, the unchanged
knowledge rate was 88.1%, the poorer knowledge rate was 3.6%. The
better practice rate was 41.7%, the unchanged practice rate was 39.3%,
the poorer practice rate was 19%.
3.2.7. The effectiveness of the intervention
Table 3.6: The effectiveness of the intervention using Generalized
Estimation Equation
Intervention

No
Yes
< 5 years

Period of time
getting DM
5-10 years
≥ 10 years

0.61

0.464

-0.80

0.74

0.280

0a
0.90

0.55

0.102

0.43

0.08

0.45

0.354

0a
0.76
0a
0.42


0.2

2.1

0.1

1.9

0.84 7.3
0.91 5.01
0.63 3.7


18
Female
BMI
Age

0a
0.003

0.025

0.905

-0.01

0.08

0.898

small rate of illiterate patients. The poor and near poor accounted for
4%. This group is more vulnerable than the others. In 2013, the rate of
near-poor households in Ha Nam was 5.37% and the poor households
was 6.28%. In our study, this rate decreased to 2.6% and 1.3%,
respectively, as our objects were mostly elderly, unlike the survey
across the whole province.
Most of the objects have health insurance (99%), far exceeding the
target of general health insurance coverage of Ha Nam province in 2017
(78.8%), as well as the rate of using health insurance in 2013 (63.8%).
It is easy to see the benefits and indispensable importance of health
insurance for patients, especially those with chronic diseases. Therefore,
it is necessary to have reasonable policies to build a sustainable health
insurance fund.


19
4.1.2. Characteristics of eye diseases
In our study, there was a complete set of diabetic eye complications.
The number of patients with high lesions (the district with the highest
number of lesion patients was Ly Nhan with 18.2%), and higher than the
study by Kawashima (2010) (10.5%).
Most of the objects had visual impairment (> 80%), which was
higher than that of Tran Thi Thu Hien (5.1%)
A general assessment of retinal damage due to diabetics revealed that
the rate of patients with damage was as high as 30.9%. Compared to the
national studies, our rate of retinal damage was higher than that of the
Institute of Endocrinology (27.8%), Nguyen Kim Luong (22.9%), lower
than the study of Nguyen Huong Thanh (33%), Tran Minh Tien (36.1%),
possibly due to previous studies having studied fewer patients and
conducted in hospitals.

Rajiv's study in India with 6.43 times higher risk of DR in people with
diabetes over 15 years.
Daniel's study (2016) also confirmed a 1% decrease in blood glucose
which reduced the risk of developing the disease by 40%. Especially for
type-2 diabetes patients, if they have good and strict blood glucose control,
more than 90% of them did not develop into hyperthyroidism.
4.2. Evaluation of the effectiveness of diabetic retinopathy
interventions
4.2.1. Intervention sites and general information
The number of study objects in each site is not different. Besides,
anthropometric and social characteristics reflect the similarity of the
study objects between the control and the intervention group at each
stage. In addition, there was no significant change in the number of
patients between the two assessments, which facilitates paired
comparisons over time in each patient group.
4.2.2. Changes in the incidence of diabetic retinopathy
Since the disease is a slow-moving disease, our study found that
the number of new cases was still low. This suggests that longer studies are
needed to monitor and assess the new incidence of DR in diabetic patients.
However, in Viet Nam, in addition to the fact that the DR is a slowmoving disease, there are no studies that can thoroughly assess the new
incidence of patients, as well as understand relationships to this status.
4.2.3. Change in eyesight status
The study results show that there was almost no change in the
visual status of objects in both intervention and comparison groups.
This is understandable because most of the interventions in our study
focus on the treatment of diabetes and prevention of retinal
complications rather than the treatment of DR. As a result, the results of
the study show that the patient's eyesight did not change, showing that
interventions have contributed to stabilizing and maintaining the
patient’s eyesight, minimizing bad progression of visual acuity over

a proliferative stage. Daniel's study (2016) confirmed that every 1% of
the blood sugar decrease would reduce the risk of developing DR by
40%. Or even this study showed that people with high blood glucose
levels are twice as likely to have DR than patients with normal blood
glucose levels.
4.2.6. Change in knowledge and practice against disease
When evaluating the knowledge, practice against disease, it
showed very good results in Binh Luc district. Patients in the intervention


22
group improved their knowledge and practice clearly after intervention
compared to that before intervention, especially the compliance with
improved disease prevention and treatment. While the patients in the
comparison group the change was completely unclear.
Positive changes in the knowledge and practice about diabetes /
DR and prevention against diabetes / DR show a remarkable effect of
interventions in health education and communication in the intervention
site, as well as improve the knowledge of patients through direct
counseling conducted by health staff during periodic exams and diabetic
follow-up visits. In addition, the intervention to even health staff also
improves professional knowledge, management skills and diabetes
monitoring. This leads to the provision of better healthcare services to
patients, as well as better management of the patient's medical conditions,
which indirectly enhances the effectiveness of the procedures of treatment
and monitoring of diabetes.
4.2.7. Assessment of the intervention effectiveness
Factors negatively affecting the incidence of DR are the poor
treatment efficacy and prolonged duration of diabetes. This finding is
consistent with many studies in the world, such as Wolfensberger (2001),