Summary of PHD thesis in medicine: Study of beta cell function, insulin sensitivity and the results from lifestyle intervention in pre-diabetes population

INTRODUCTION

β

Prediabetes   (or   “intermediate   hyperglycaemia”),   based   on   glycaemic  parameters above normal but below diabetes thresholds is a high risk state  for diabetes with an annualized conversion rate of 5%–10%; with similar  proportion   converting   back   to   normoglycaemia.   The   prevalence   of  prediabetes is increasing worldwide and it is projected that >470 million  people will have prediabetes in 2030. Prediabetes is associated with the  simultaneous   presence   of   insulin   resistance   and     ­cell   dysfunction, abnormalities   that   start   before   glucose   changes   are   detectable.   For  prediabetic individuals, lifestyle modification is the cornerstone of diabetes  prevention   with   evidence   of   a   40%–70%   relative   risk   reduction.  Accumulating data also suggests potential benefits from pharmacotherapy.

In order to understand more about diabetes intervention in Vietnamese,  a research entitle: “Study of beta cell function, insulin sensitivity and the  results   from   lifestyle   intervention   in   pre­diabetes   population”,  has   been  conductedwith two objectives:

1. To investigate risk factors, beta cell function, insulin sensitivity

among pre­diabetes in Ninh Binh.

2. To evaluate the outcome of lifestyle modification in pre­diabetes

after 24 months.

Essential of the thesis: A better knowledge of  beta cell function, insulin sensitivity  and pre­ diabetes prevention should provide important insights in the pathogenesis  of   diabetes,   allowing   the   evaluation   of   better  interventions  at   both  population level and patient level to reduce the burden of diabetes.

The new main scientific finding of the PhD thesis: - This is the first research in Vietnam to study the changes in beta cell

function, insulin resistance following lifestyle intervention in pre­diabetes.

- Base on impaired beta cell function, insulin resistance can predicting

the outcome of devenlopment diabetes after 24 months.

Thesis Layout: The thesis with 120 pages, include: introduction(2 pages), Chapter 1:  Literature   Review   (33   pages),   Chapter   2:   Subjects   and   methodology(25  pages), Chapter 3: Results (27 pages), Chapter 4: Discussion (22 pages),  Conclussion (2 pages), Recomendation (1 page).

The thesis has 153 references (37 in Vietnamese: 116 in English).

CHAPTER 1: LITERATURE  REVIEW

2.1. PRE­DIABETES AND RISK FACTORS

1.1.1. History and concept of pre­diabetes

Diabetes is   one   of   the   first   diseases   described with   an   Egyptian  manuscript   from c. 1500 BCE mentioning   “too   great   emptying   of   the  urine”. The first described cases are believed to be of type 1 diabetes. The  term   "diabetes"   or   "to   pass   through"   was   first   used   in   250 BC   by   the  Greek Apollonius of Memphis.

Prediabetes means that your blood sugar level is higher than normal  but not yet high enough to be type 2 diabetes. Without lifestyle changes,  people with prediabetes are very likely to progress to type 2 diabetes. 1.1.2. Epidemiology of pre­diabetes

The International Diabetes Federation (IDF)estimates that the number of  people with pre­diabetes is 316 million (6.9%) in 2013, and it will increase about  471 million (8.0%) by 2035. In Vietnam (2001), one national study of diabetes  prevalence, reported a prevalence of pre­diabetes of 5.9% 1.1.3. Diagnosis and classification of diabetes, pre­diabetes.

Diagnostic   methods:   Capillary   Blood   Glucose;   Oral   Glucose  Tolerance   Test;   HbA1c;   Diagnosis   criteria   of   diabetes,   pre­diabetes   of  WHO­IDF 2010. 1.1.4. Risk factors of pre­diabetes

Diabetes   risk   factors:   Age;   race;   history   of   diabetes   family;   blood  glucose   disorder;   overweight/obesity,   hypertension;   gestational   diabetes;  inactivity; smoking, stress. 1.1.5. Metabolic syndrome and pre­diabetes

"Metabolic syndrome"  is a  group of  metabolic disorders associated  with   risk   factors   for   cardiovascular   disease   as   well   as   predicting   the  development of diabetes. 1.1.6. Dyslipidemia and pre­diabetes

CELL   FUNCTION,   INSULIN   SENSITIVITY   IN   PRE­

Insulin resistance causes abnormal lipid metabolism inbody tissues. In  contrast, visceral adipose deposits release adipocytokines that contribute to  insulin resistance. 2.2.

β DIABETES

1.2.1. Insulin sensitivity, insulin reristance

Insulin   resistance   is   a   decrease   in   the   biological   response   to   both  endogenous and exogenous insulin. Insulin resistance exists in both the liver  and peripheral tissues, which is manifested by a decreased ability to inhibit  glucose production in the liver, using glucose in peripheral tissues. 1.2.2. Beta cell function

Beta   cell   fucntion   defect   is   involved   with   many   metabolic,   vascular  abnormalities, esspecially the pathogenesis of type 2 diabetes. Studies indicated  that over 50% of new onset diabetes with impaired beta cell function. 1.2.3. Impaired cell fucntionand insulin resistance

IGT and IFG are intermediate states in glucose metabolism that exist  between normal glucose tolerance and overt diabetes. Insulin resistance and  impaired   beta­cell   function,   the   primary   defects   observed   in   type   2  diabetes, both can be detected in subjects with IGT and IFG. 1.2.4. Assessment of beta cell function, insulin resistance

Methods for evaluating insulin resistance, beta cell function include  measuring a fasting insulin, a fasting blood sugar, and an adiponectin value  and using an insulin index, HOMA1, HOMA2,… 2.3. PRE­DIABETES PREVENTION

The last two decades have seen an explosive increase in the number of  people with diabetes globally. There is now an urgent need for strategies to  prevent   the   emerging   global   epidemic.   Several   recent   successful  intervention studies, both lifestyle and pharmacological, targeting subjects  with IGT have stimulated enthusiasm for prevention of Type 2 diabetes.  Lifestyle interventions reduced the incidence of diabetes by over 50% in  the   Finnish   Diabetes   Prevention   Study   and   the   Diabetes   Prevention  Program.   Longer­term   follow­up   studies   with   both   lifestyle   and  pharmacological   interventions   actually   showed   the   efficacy   of   diabetes  prevention or merely delay its onset.

CHAPTER 2: SUBJECTS AND METHODOLOGY

2.1. SUBJECTS 2.1.1. The Subjects of the research:

+ Totally 757 subjects were divided into 2 groups: Group of 618 pre­diabetes  (220 males and 398 females), ages 30­75 and control group of 139 nomal people at  the same ages and living the same areas with group of pre­diabetes.

+ In order to analyze the data, group of pre­diabetes were divided into

two group (compliance groups and non­compliance groups). 2.1.2. Selection and Exclusion criteria 2.1.2.1. * Selection critetia for risk factor subjects

Over   weight/obesity,   inactivity;   Hypertention;   Dyslipidemia;   age>

45s; History of Diabetes family,… 2.1.2.2. * Selection critetia for pre­diabetes

Sreened from people with risk factors; Age between 30 to 75 years;

2.1.2.3. * Selection critetia for normal people

Age between 30 to 75 years; non­obese with a BMI < 23kg/m2, waist  circumference of whom male <90cm & female <80cm, non­diabetic, non­ hypertensive,   those   without   family   history   of   type   2   diabetes,  normolipidemic and is a non­heavy­smoker. 2.1.2.4. * Groups division

Compliance  groups:   carry  out  the  full  range  of   interventions.  Con­

compliance groups: lack of carrying out the full range of interventions. 2.1.2.5. Exclusion criteria

Subjects suffering from chronic illness, ketosis, chronic liver and renal  diseases,   pregnant   women,   Subjects   taking   anti­hypertensive   drugs,  steroids   or   hormonal   products   were   excluded   from   the   study.   Subjects  taking anti­hypertensive drugs, steroids or hormonal products. 2.1. METHODOLOGY 2.1.1. Methods

It was a cross – sectional, descriptive study with control group and

case control study. 2.1.2. Sampling:

+ Sample size for objective 1: calculated on the basis of the sample

size of the descriptive study.

+ Sample size for objective 2: based on the purpose of the research to  evaluate   the   outcome   of   lifestyle   modification   (the   effectiveness   of  interventionseffects) on beta­cell function and insulin sensitivity. 2.1.3. Study period and site

Study period from October 2011 to December 2015 and 24 months of

intervention. Study site: Ninh Binh and Tam Diep City Ninh Binh.  2.1.4. Research indicators

Medical   history,   pre­history;   physical   examination,   co­morbidities,  smoking habits, Body weight, Pulse and blood pressure, The fisrt Blood  specimen were assayed glucose, insulin, C­peptid, lipid levels. 2.1.5. Collection   of   anthropometric   measurements   and   blood  specimens

Interviews, pre­history; physical examination, co­morbidities, habits,  Body   weight,   Pulse   and   BP,   The   fisrt   Blood   specimen   were   assayed  glucose, insulin, C­peptid, lipid levels.

version   2.2.,

available

Insulin sensitivity, beta cell fucntion, insulin resistance was calculated  using   HOMA   Calculator from  http://www.ocdem.ox.ac.uk/. We used paire of fasting plasma glucose (in  mmol/l) and RIA insulin (in pmol/l) concentrations.

2.1.6. Research content, process, approach 2.1.6.1. Screening high risk people and pre­diabetes

Interviews,   medical   history,   pre­history;   physical   examination,   co­

morbidities, OGTT, fasting glucose insulin, lipid. 2.1.6.2. Intervetion and follow­up

Education   sessions   of   diabetes   prevention;   Lifestyle   modification;  follow­up   and   councelling   were   carried   in   a   three­month  interval in the first year and in a six­month interval in the 2nd.

2.1.6.3. Data management and monitoring

Data were collected at baseline and 24th months.

2.1.7. Criteria for diagnosis

(cid:0) Diabetes, prediabetes: WHO­IDF 2010. (cid:0) Metabolic Syndrome : IDF and NCEP ATPIII. (cid:0) Dyslipidemia:  EGIR/ESC (cid:0) Waist, Hip, IR and impaired beta­cell function: WHO

2.1.8. Methods 2.1.9. Analytical methods

Managing, processing data by Epidata 3.0, SPSS 16.0 software.

2.1.10. Research Ethics

The study did not do any harm or danger to patients.  Ethical committee

approval had been obtained 24/5/2017, IRB No 87/HĐĐĐ­ĐHYHN.

3. CHAPTER 3: RESULTS

2.2. General characteristics of participants 2.2.1. Characteristics of age, sex, and phisical index

Table 3.­3.2:Proportion of sex and average age of the study population

p Chara cterist ics Pre­ diabet es  (n=61 8) Contr ol  group  (n=13 9)

<0.01 57.2 ±  8.5 55.2 ±  8.5

Mean  age(ye ar) Gender Sex Male n 220 % 35.6 n 52 % 37.4 >0.05

Average   age   of   pre­diabetes   was   higher   than   control   group   (p  <0.01); the percentage of male was higher than female; the proportion of  subjects in both groups was from 50 to 69 years old. 2.2.2. Anthropometric characteristics of the study population

Female Total 398 618 64.4 100 87 139 62.6 100

Table 3.3­3.: Comparison of anthropometric characteristics between pre­diabetes  group and control group Pre­diabetes  (n=618) 23.1 ± 2.7

Characteristics p

BMI (kg/m2)

Waist(cm)

Hip(cm) Total Male Female Total Male Female 79.9 ± 7.9 82.8 ± 7.9 78.2 ± 7.5 89,6 ± 7,9 90.8 ± 7.5 88.9 ± 8.1

diabetes group were significantly higher than that in the control group.

Control group  (n=139) 21.8 ± 1.4 75.9 ± 7.7 77.9 ± 11.1 74.9 ± 4.4 87.5 ± 5.6 88.0 ± 5.6 87.2 ± 5.6 121.3 ± 7.7 75.8 ± 9.3 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 144.6 ± 21.2 82.7 ± 12.9 SBP (mmHg) DSP (mmHg) The average index of WC, HC, BMI, HR, SBP, and DBP in pre­

2.2.3. Biochemical characteristics

Table 3.: Comparisons of blood biochemical characteristics among study groups

Characteristics p

The   average   index   of  glucose,   gluocse  2h,   lipid   among  pre­

diabetes were significantly higher than that in the control group. 2.3. Risk   factors   and   insulin   resistance,   insulin   sensitivity,   beta   cell

function

Glucose (mmol/l) Glucose 2h (mmol/l) TC (mmol/l) TG (mmol/l) HDL­C (mmol/l) LDL­C (mmol/l) Pre­diabetes  (n=618) 5.95 ± 0.54 8.17 ± 1.37 5.86 ± 1.12 2.20 ± 1.73 1.28 ± 0.28 4.49 ± 2.11 Control group  (n=139) 5.49 ± 0.34 7.01 ± 0.67 4.65 ± 0.57 1.38 ± 0.47 1.26 ± 0.30 2.77 ± 0.54 <0.01 <0.01 <0.01 <0.01 >0.05 <0.01

2.3.1. Risk factors among pre­diabetes

Table 3.­3.7: Phisical activities and habit factors of pre­diabetes

p Total (n=618 ) Male (n=220 ) Risk  Factor s Femal e (n=398 ) n n % % n %

73 11.8 73 33.2 0 <0.01

197 31.9 161 73.2 36 9.0 <0.01

80 12.9 32 14.5 48 12.1 >0.05

75 12.1 29 13.2 46 11.6 >0.05 Smoki ng Alcoho lic  Drink Lack  of  eating  fruits/v egetabl es Oil/fat  often  use

Work 107 309 17.3 50.0 47 116 15.1 48.5 <0.05 >0.05 vigorous Moderate 376 60.8 125 56.8 21.4 52.7 251 60 193 63.1 >0.05 Travel

(Bicycl e) Phisical  activities 55 242 vigorous Moderate <0.05 >0.05 8.9 39.0 12.7 41.8 6.8 37.7 27 150

Figure 3.: The rate of overweight, obesity by gender among pre­diabetes The rate of overweight, obesity was 50,1%; there wasn’t difference

between males and female.

Table 3.: Percentage of abdominal obesity by sex in people with pre­diabetes

28 92 Prevalence of prediabetes who smoke, drink alcohol was very high;  Work vigorously and moderately was  67.3%;  did vigorous and moderate  phisical activities was 47.9%.

p Total (n=61 8) Male (n=2 20) Inde x

%

<0.01 Wai st

In people with pre­diabetes, percentage of abdominal obesity in men

was higher than women (p<0.05).

Figure 3.: Proportion of hypertention (JNCVII) among pre­diabetes

Prevalence   of   hypertension   (JNCVII)was  36.1%;  pre­hypertention  was  53.4%. There was difference  in prevalence of hypertension between  men and women.

Table 3.: Proportion of dyslipidemia among pre­diabetes population

W/H <0.05 n Nomal Elevated  Nomal Elevated 412 206 229 389 Fema le (n=3 98) n 66.7 33.3 37.1 62.9 % 232 166 136 262 % 180 40 93 127 n 81.8 18.2 42.3 57.7 58.3 41.7 34.2 65.8

p Total (n=61 8) Male (n=22 0) Dyslip idemia Femal e (n=39 8)

n % n % % n

TC 464 75.1 163 74.1 75.6 >0.05 301

TG 308 49.8 129 58.6 45.0 <0.01 179

HDL­C 22 3.6 13 5.9 9 2.3 <0.05

LDL­C 489 79.1 172 78.2 317 79.6 >0.05

The   incidence   of   dyslipidemia   was   89.8%.   Rate   of   disorder  decreased from LDL­C, TC, TG and HDL­C respectively; The prevalence  of TG and HDL­C disorders in men is higher than that of women (p <0.01).

Figure 3.: Prevalence of metabolic syndrome in pre­diabetic individuals

Prevalence   of   metabolic   syndrome  (ATPIII   criteria)   was   higher  than IDF criteria. Prevalence of MS in female was higher than that in male  (p<0.01).

Table 3.:  Summary of some diseases, syndrome in patients with pre­diabetes

Dyslipidemia 575 89.8 200 90.9 355 89.2 >0.05

p Total (n=61 8) Male (n=22 0) BMI  index

Femal e (n=39 8) n n % % n %

310 50.1 117 53.2 193 48.5 >0.05

Over  weight /obesit y

AO WC WC/HC 206 389 33.3 62.9 40 127 18.2 57.7 166 262 41.7 65.8 <0.05 <0.05

223 36.1 96 43.6 134 33.6 <0.05

90.9 575 89.8 200 355 89.2 >0.05 Hypert ension  degree  1,2 Hypert ension

In   pre­diabetes:   Prevalence  of   dyslipidemia  was  89.8%;  Over   was  62.9%;   Abdomal   obesity

(AO)

weight/obesity  was  50.1%; Hypertension (JNCVII) was 36.1%. 2.3.2. Insulin resistance, beta cell function, insulin sensitivity 2.3.2.1. Insulin resistance, beta cell function, insulin sensitivity

Table 3.­3.12: Comparison of the limit values of insulin, HOMA2 index

HCCH 29.8 51.9 184 321 IDF ATPIII 33  111 15.0 50.5 151 210 37.9 52.8 <0.05 >0.05

In   pre­diabetes   group,   insulin   index,   HOMA2­IR   index   was  higher;   conversely,   the   HOMA2­%S   was   lower   than   the   control   group;  There was no difference in HOMA2­% B index.

Figure 3.: Insulin resistance rate according to insulin levels in pre­diabetes group

There   wasn’t   difference   in   the   rate   of  pre­diabetes   with  insulin

resistance between men and women. Table 3.:  The rate of insulin resistance, impaired beta cell function, impaired insulin   sensitivity by HOMA2 in pre­diabetes

Index

p

Male (n=220)

Female (n=398)

HOMA2­IR

HOMA2­%S

HOMA2­%B

Total (n=618) Normal 325(52.6) Elevated     293 (47.4) 276 (44.7) Inpaired 342 (55.3) Normal 439 (71.0) Inpaired 179(8.9) Normal

111(50.5) 109 (49.5) 104 (47.3) 116 (52.7) 164 (74.5) 56(25.5)

214 (54.0) >0.05 184 (46.0) 172 (43.2) >0.05 226 (56.8) 275 (69.1) >0.05 123(0.9)

The rate of insulin resistance, impaired beta cell function, impaired

Equation

Index

insulin sensitivity in pre­diabetes ranges from 44.7% to 71%. 2.3.2.2. The relationship, correlation between insulin resistance, beta cell   function, insulin sensitivity with risk factors Table 3.: Correlation between insulin resistance, impaired beta cell function, impaired   insulin sensitivity with age in pre­diabetes r 0.014 0.017

y =  ­ 0.04 x  +   47.09   y =  ­ 0.80 x  +   0.93

Insulin HOMA2­IR

p >0.05 >0.05

Index p Pre­ diabetes  (n=618) Control  group  (n=139) (?+SD) <0.01 Insulin LV (?+SD) <0.01 HOMA2­IR LV (?+SD) <0.01 HOMA2­%S LV (?+SD) >0.05 HOMA2­%B LV 45.1±21.5 20.0­319.5 0.88±0.41 0.38­5.78 132.6±50.3 17.3­259.1 60.5±21.6 26.3­244.8 37.8 ± 31.6 10.76­258.5 0.72 ± 0.58 0.20­4.76 183.1 ± 85.1 21.0­483.7 60.43±29.47 27.4­240.0

HOMA2­%S HOMA2­%B

y =    0.24 x  +   119.00   y =  ­ 0.05 x  +   63.46

0.041 0.020

>0.05 >0.05

In   pre­diabetes,   there   was   no   correlation   between   insulin

resistance, beta cell function, insulin sensitivity with age (p> 0.05).

Table 3.: Comparison of the values of insulin resistance, insulin sensitivity, beta cell   function by gender in pre­diabetes Pre­diabetes  (n=618) Male (n=220) 44.13 ± 17.63

p Index

Female (n=398) 45.65 ± 23.48 Insulin <0.01

HOMA2­IR 0.86 ± 0.33 0.89 ± 0.41 >0.05

In pre­diabetes, there was a difference in HOMA2­%B between  men and women. There was no difference in HOMA2­% S, HOMA2­IR in  the pre­diabetes by gender.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with BMI index in pre­diabetes.

Index

p

Insulin

BMI<23 kg/m2 (n=304) 42.66 ± 16.43

BMI≥23 kg/m2 (n= 314) 47.48 ± 25.40

<0.01

HOMA2­IR

0.83 ± 0.31

0.92 ± 0.47

<0.01

HOMA2­%S

136.89 ± 49.39

128.62 ± 51.07

<0.01

HOMA2­%B

58.97 ± 19.62

62.03 ± 23.45

>0.05

There   was   a   correlation   between   insulin   index,   HOMA2­IR,  HOMA2­%S with BMI index. There is no correlation between beta cell  function with BMI index.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with hypertension in pre­diabetes

Index

p

Insulin

Hypertesion  (n=370) 45.34 ± 21.58

Fre­Hypertension (n=248) 44.76 ± 18.31

>0.05

HOMA2­IR

0.88 ± 0.44

0.87 ± 0.35

>0.05

HOMA2­%S

132.82 ± 50.62

132.48 ± 50.12

>0.05

HOMA2­%B

60.59 ± 22.86

60.43 ± 19.86

>0.05

There   was   no   relationship   between   insulin   resistance,   insulin

sensitivity, beta cell function with hypertension in pre­diabetes.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with abdominal obesityin pre­diabetes Without (n=412)

With (n=206)

Index

p

133.91 ± 52.29 132.01 ± 49.34 >0.05 HOMA2­%S HOMA2­%B 57.69 ± 19.21 62.09 ± 22.82 <0.05

Insulin

51.64 ± 28.59

41.84 ± 16.11

<0.01

HOMA2­IR

1.00 ± 0.532

0.817 ± 0.310

<0.01

HOMA2­%S

119.07 ± 47.54

139.49 ± 50.43

<0.01

HOMA2­%B

66.77 ± 45.54

57.40 ± 18.66

<0.01

HOMA2­IR,   HOMA2­%B   index   in   pre­diabetes   group   with  abdominal   obesity   was   higher   than   in   non­abdominal   obesity   group.  Conversely, HOMA2­% S was lower than non­abdominal obesity group.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with lipid disorders in pre­diabetes

Index

Dyslipidemia (n=555) Without (n=63)

p >0.05 >0.05 >0.05 >0.05

Insulin HOMA2­IR HOMA2­%S HOMA2­%B

45.31 ± 21.76 0.88 ± 0.41 132.44 ± 50.92 61.02 ± 22.22

43.29 ± 20.04 0,85 ± 0.39 134.82 ± 45.60 56.15 ± 15.89

There was a difference in insulin  index, HOMA2­%B, HOMA2­%S,

HOMA2­IR between group with and without dyslipiemia.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with MS (ATPIII) in pre­diabetes p

MS (ATPIII)

Index

<0.01 >0.05 >0.05 >0.05

Insulin HOMA2­IR HOMA2­%S HOMA2­%B

With (n=321) 46.57 ± 23.99 0.91 ± 0.45 129.20 ± 50.37 59.85 ± 22.21

Without (n=297) 43.53 ± 18.53 0.85 ± 0.36 136.45 ± 50.19 61.25 ± 21.13

In pre­diabetes  with MS, insulin  index was higher than in group  without   MS;  there   was  no  difference   in   HOMA2­%B,   HOMA2­%S,  HOMA2­IR between groups with and without MS.

Table 3.: The relationship between insulin resistance, insulin sensitivity, beta cell   function with MS (IDF) in pre­diabetes p

Index

Insulin HOMA2­IR HOMA2­%S HOMA2­%B

MS ( IDF) With (n=184) 51.72 ± 28.89 1.01 ± 0.54 118.69 ± 47.65 65.64 ± 24.66

Without (n=434) 42.31 ± 18.86 0.83 ± 0.32 138.62 ± 50.38 58.36 ± 19.94

<0.01 <0.01 <0.01 <0.01

HOMA2­IR, HOMA2­% B in groups with MS (IDF) was higher;

HOMA2­%S was lower than group without MS.

Table 3.: Comparison of insulin resistance, insulin sensitivity, beta cell function  among sub forms of pre­diabetes

Index p

Insulin HOMA2­IR HOMA2­%S HOMA2­%B IFG (n=153) 41.0 ± 15.8 0.81 ± 0.31 139.2±48.6 49.8±15.3 IGT (n=322) 46.2 ± 23.5 0.89 ± 0.43 131.6±49.59 69.54±22.19 IFG&IGT (n=143) 47.1 ± 21.9 0.94 ± 0.43 127.9±53.5 51.6±17.1 <0.05 <0.05 <0.05 <0.01

Different forms of pre­diabetes vary in insulin index, HOMA2­IR,  HOMA2­%S, HOMA2­%B; Insulin resistance, impaired insulin sensitivity  increased from IFG, IGT, and IFG & IGT.

Table 3.: Comparison of the rate of insulin resistance, insulin sensitivity, beta cell   function among sub forms of pre­diabetes

Index p IFG  (n=153) IGT  (n=322)

Insulin <0.01

HOMA2­IR >0.05

HOMA2­%S >0.05

The   rate   of   insulin   resistance,   impaired   insulin   sensitivity,

impaired beta cell function decreased from IGT, IGT&IFG and IFG. 2.4. The Results of intervention, treatment 2.4.1. Changes in risk factors after intervention

Figure 3.: Changes in risk factors  in pre­diabetes after 24 months Interventions   reduced  the   rate   of   eating   less   vegetables   and  regularly eat fat; However, the rate of smoking, alcohol drink changed very  little.

Table 3.: Comparison of lifestyle changes after 24 months of  intervention

HOMA2­%B <0.01 N H N H L N N L IFG&IG T (n=143) 245(49.7) 77(61.6) 178 (41.9) 144 (49.3) 139 (50.4) 183 (53.5) 177 (40.3) 145 (81.0) 136(27.6) 17(13.6) 85(19.9) 68(23.3) 61(22.1) 92(26.9) 140(31.9) 13(7.3) 112(22.7) 31(24.8) 63(14.9) 80(27.4) 76(27.5) 67(19.6) 122(27.8) 21(11.7)

p Non­ C (n=295) Habit Complia nce (n=323)

n 26 103 % 8.1 32.2 n 30 95 % 10.5 33.2 >0.05 >0.05

39 12.2 62 21.7 <0.01

The rate of smoking, drinking alcohol  without changes; however  the   results   showed   that   decrease   in   the   rate   of   lack   of   eating  fruits/vegetables.

Figure 3.: Changes in phisical activities after 24 months of intervention

Smoking Alcohol drink Lack   of   eating  fruits/vegetables Oil/fat often use 18 5.6 33 11.5 <0.05

Vigorous  work decreased; The average rate  of  moderate  phisical

activities (P.A), cicling increased after 24 months of intervention.

Table 3.: Comparison of phisical activities after 24 months between group of

compliance and without compliance

p Non­ C (n=295)

Phisica l  activiti es % n %

Work Compli ance (n=323) n vigorous Moderate 39 207 12.2 64.7 40 156 14.0 54.5 >0.05 <0.05

251 78.4 200 69.9 <0.05

Travel  (Bicycl e) Phisical  activities vigorous Moderate <0.05 <0.01 5.2 43.0 15 123 33 191

Table 3.: Changes in anthropometric index after 24 months of intervention

10.3 59.7 In the compliance group, the rate of people with pre­diabetes did  moderate works, moderate phisical activities and traveling by bycircle was  statistically higher than in statistically the non­compliance group.

Complian ce p Phisical  index Non­ C (n=295)

BMI (n=323) 23.1 ± 2.7 23.0 ± 2.8 >0.05

Waist

The   results   showed   no   difference   statistically   in   Weight,   BMI,

WC, HC index after intervention.

Figure 3.7: Comparison of changes in overweight and obesity rates after 24 months

The prevalence of overweight and obesity of the compliance group

was significantly lower than that of the non­compliance group. 2.4.1.1. Changes in blood lipids

Table 3.: Changes in lipid after 24 months of intervention in pre­diabetes p

Hip Male Female Male Female 82.9 ± 7.8 78.2 ± 7.5 90.8 ± 7.5 88.9 ± 8.0 82.7 ± 8.3 78.8 ± 7.5 91.7 ± 7.1 90.4 ± 7.6 >0.05 >0.05 >0.05 >0.05

24 months (n=618) Baseline (n­618) Index

The   mean   of   TC,   LDL­C   index   decreased   significantly;

Conversely, the mean of TG, HDL­C index increased significantly.

Table 3.: Comparison of lipid profiles after interventions between the two groups

Cholesterol Trilycerit HDL­C LDL­C 5.86 ± 1.13 2.20 ± 1.73 1.12 ± 0.261 3.62 ± 1.95 5.07 ± 1.00 2.48 ± 2.34 1.28 ± 0.28 2.72 ± 1.34 <0.01 <0.01 <0.01 <0.01

Index p

There   was   no   different   in   TC,   TG,   LDL­C   index   between  compliance   and   non­compliance   group.   HDL­C   index   in   the   non­ compliance group was significantly lower than the compliance group.

Table 3.: Change in the rate of dyslipidemia after intervention

Cholesterol Trilycerit HDL­C LDL­C Compliance (n=323) 5.03 ± 0.92 2.46 ± 1.97 1.15 ± 0.30 2.95 ± 1.59 Non­ C (n=295) 5.11 ± 1.08 2.49 ± 2.70 1.09 ± 0.22 2.98 ± 1.01 >0.05 >0.05 <0.01 >0.05

Dyslipidemia p

After intervention, the prevalence of pre­diabetes with TC, LDL­C  disorder and dyslipidemia decreased significantly; conversely, the rate of  TG, HDL­C disorder increased significantly.

Figure 3.8: Comparison of changes in the rate of lipid disorders after 24 months  between the compliance group versus non­compliance group

After 24 months of intervention, there was no  difference in the  prevalence of lipid disorders between the compliance group and the non­ compliance group. 2.4.2. Changes in blood glucose after intervention

Table 3.: Changes in blood glucose levels in pre­diabetes group after intervention

Cholesterol (≥5,2) Trilycerit (≥1,7) HDL­C (<0.9) LDL­C (≥3,4) Dyslipidemia Compliance (n=323) 464(75.08) 308(49.84) 22(3.56) 489(79.13) 555(89.81) Non­ C (n=295) 263(42.56 351(56.79) 75(12.14) 177(28.64) 456(73.79) <0.01 <0.01 <0.01 <0.01 <0.01

Index p

Glucose (mmol/l) Baseline (n­618) 5.94 ± 0.54 24 months  (n=618) 5.51 ± 0.95 <0.01

After   intervention,   fasting   glucose   index,   glucose   2h   decreased

significantly compared to baseline values.

Glucose 2h (mmol/l) 8.17 ± 1.37 7.98 ± 2.92 <0.01

Index p Compliance (n=323)

Baseline 24th >0.05 <0.01 Glucose  (mmol/l)

Table 3.: Comparison of changes in mean blood glucose in two groups Non­ C (n=295) 5.94 ± 0.55 5.36 ± 0.78 <0.01 8.14 ± 1.33 7.76 ± 2.64 <0.01

After   intervention,   fasting   blood   glucose,   glucose   2h   in   the

compliance group decreased and lower than that in non­compliance group. Table 3.: Comparison of the risk of developing diabetes after 24 months

p Baseline 24th >0.05 <0.05 Glucose 2h (mmol/l) 5.96 ± 0.54 5.68 ± 1.09 <0.01 8.22 ± 1.42 8.22 ± 3.19 >0.05 p

p Status at  baseline Sau CT IFG(n=153) IGT(n=322) IFG&IGT(n=143)

DM 16(10.5) 29(9.0) 31(21.7) <0.01

Without DM 137(89.5) 293(91.0) 112(78.3)

the lowest rate of diabetes, after 24 months.

Figure 3.9: Comparison of rates of status of diabetes, pre­ diabetes, normal glucose

The incidence of diabetes in the non­compliance group was higher

than the compliance group. 2.4.3. Changes in beta cell fucntion, insulin sensitivity, IR

Table 3.: Changes in beta cell fucntion, insulin sensitivity, insulin resistance

<0.01 IFG & IGT group had the highest rate of diabetes; IGT group had

p Index

Insulin HOMA2­IR HOMA2­%S HOMA2­%B Status after  24 months NG (n=308) 45.7±26.9 0.87 ± 0.51 164.2±121.6 83.3±35.8 Pre­DM(n=222) 54.9±42.2 1.06 ± 0.78 143.7±113.1 75.08±36.10 DM (n=76) 63.38 ± 67.92 1.24 ± 1.27 136.9±104.8 55.0± 26.4 <0.01 <0.01 <0.05 <0.01

Normal glucose group, insulin index, insulin resistance in normal  group was lowest; beta cell fuction and insulin sensitivity was higher than  those in pre­diabetes and diabetes group (p <0.01).

Table 3.: Changes in beta cell fucntion, insulin sensitivity, insulin resistance after 24  months of intervention Baseline  (n­618) 45.11  ± 21.58 0.88 ± 0.41 132.69 ± 50.38  60.52 ± 21.69

Index p

After intervention, the mean of insulin index, insulin resistance,

beta cell fuction and insulin sensitivity index increased significantly. Table 3.: Comparison of  insulin index after 24 months between compliance and non­ compliance group

24 months  (n=618) 51.32 ± 40.36 0.98 ± 0.76 153.31 ± 116.94 76.76 ± 36.00 Insulin HOMA2­IR HOMA2­%S HOMA2­%B <0.01 <0.01 <0.01 <0.01

Index p Non­C (n=295) Complianc e (n=323)

Insulin in compliance group didn’t increase; there was diference

statistically between compliance group campare with non­compliance.

Figure 3.10: Comparison of the rate of insulin resistance

After intervention, the incidence of hyperinsulinemia in the non­

compliance group was significantly higher than the compliance group.

Table 3.: Comparison of mean HOMA2­IR index after intervention

Insulin Baseline 24th >0.05 <0.01 p 44.89 ± 20.23 46.35 ± 30.80 >0.05 45.35 ± 22.99 56.89 ± 48.37 <0.01

Index p Compliance (n=323) Non­C  (n=295)

HOMA2­IR   index   in   the   compliance   group   was   unchanged  compared   with   the   group   of   non­compliance   increased   significantly,   p  <0.01; HOMA2­IR index was higher than the compliance group.

Figure 3.11: Comparison of HOMA2­IR after intervention between two groups

HOMA2­IR Baseline 24th >0.05 <0.01 p 0.87 ± 0.39 0.89 ± 0.55 >0.05 0.88 ± 0.43 1.09 ± 0.92 <0.01

After   intervention,   the   HOMA2­IR   index   (insulin   resistance)   in   the  compliance group was lower significantly than the non­compliance

Table 3.: Comparison of HOMA2­% S between two groups

Index p Non­C  (n=295)

The   mean   HOMA2­%   S   (insulin   sensitivity)   index   in   the

compliance group was higher than the non­compliance group.

Figure 3.12: Comparison of the rate of patients by HOMA2­% S after intervention in   compliance groups versus non­compliance groups

The rate of decrease in insulin sensitivity in the compliance group  was   lower   than   the   non­compliance   group.   HOMA2­%   S   (insulin  sensitivity) increased was higher than the non­compliance group.

HOMA2­%S Compliance (n=323) Baseline 24th >0.05 >0.05 p 135.2 ± 53.4 161.6±120.2 <0.01 129.9 ± 46.8 143.9±112.5 <0.05

Index p Compliance (n=323)

Table 3.: Comparison of beta fucntion by HOMA2­% B between compliance and   non­compliance group Non­C  (n=295) 60.63 ± 22.21 76.65 ± 35.31

HOMA2­%B  index in  both  groups  increased,  and there  was  no

difference between the two groups after intervention (p> 0.05).

Figure 3.13: Comparation of the rate of patients according to HOMA2­% S after  intervention between compliance and non­compliance group

After intervention, there was no difference between the two groups.  Table 3.: The risk of developing diabetes after 24 months according to insulin index at  baseline

HOMA2­%B 60.40 ± 21.14 76.89 ± 36.83 Baseline 24 th >0.05 >0.05

DM n(%) p Status at  baseline OR (95%­CI) Without  DM n  (%)

After 24 months, group with elevated insulin was 1.39 times more

to develop diabetes (CI 95% CI 0.78­2.45).

Insulin High 19(15.7) 102(84.3) >0.05 1.39  (0.78­2.45)

Table 3.:  Risk of developing diabetes after  24 months according HOMA2 index

p

>0.05

>0.05

resistance group was 1.3 times more (95% CI 0.81­2.11).

Table 3.: The risk of developing diabetes after 24 months based on insulin and   HOMA2 index between compliance group and the non­compliance group

>0.05 Status at  baseline HOMA2­ IR HOMA2­ %S HOMA2­ %B 1.30 (0.81­2.11) 1,21 (0.75­1.95) 0.99 (0.58­1.67) Diabetes  n(%) High N Impaired N Impaired N Normal  n(%) 40(14.1) 36(11.2) 40(13.6) 36(11.5) 54(12.5) 22(12.5) OR (95%­CI) 244(85.9) 286(88.8) 254(86.4) 276(88.5) 378(87.5) 152(87.4) After   24   months,   the   risk   of   developing   diabetes   in   insulin

DM in Non­compliance (n=323) DM in  Complian ce (n=323) Status   at  baseline n(%) n(%) OR(95%­CI)

Insulin

High N High BT

than that in non­compliance group.

0.69 (0.25­1.88) 1.01 (0.47­2.13) 1.76 (0.55­2.49) 1.00 (0.44­2.27) HOMA2 ­IR HOMA2 ­%S HOMA2 ­%B OR(95%­ CI) 5(7.1) 25(10) 14(9.4) 16(9.4) Impaired 16(10.1) 14(8.7) N 21(9.4) Impaired 9(9.4) N 14(27.5) 32(13.6) 26(19.3) 20(13.2) 24(17.8) 22(14.6) 33(15.9) 13(16.7) 2.40 (1.17­4.93) 1.56 (0.82­2.95) 1.27 (0.67­2.39) 0.94 (0.47­1.90) The risks of developing diabetes in the compliance group was lower

4. CHAPTER 4: DISCUSSION

2.5. General characteristics of participants

Age and  sex:  Average  age of  pre­diabetes was higher than  control  group   (p   <0.01);   the   percentage   of   male   was   higher   than   female;   the  proportion of subjects in both groups was from 50 to 69 years old. (Table   3.1), (Table 3.2).

Anthropometric characteristics: The average index of WC, HC, BMI,  HR, SBP, and DBP in pre­diabetes group were significantly higher than  that in the control group (Table 3.3).

Biochemical characteristics: The average index of glucose, gluocse2h,

lipid among pre­diabetes were significantly higher than that in the control  group (Table 3.5). 4.2. Risk factors and insulin resistance, insulin sensitivity, beta cell function 4.2.1. Risk factors among pre­diabetes

Prevalence of prediabetes who smoke, drink alcohol was very high; Work  vigorously and moderately was  67.3%;  did vigorous and moderate phisical  activities was 47.9%. Cao My Phuong, Dinh Thanh Hue showed that the rate  of   smoking   was  22.2%  and   drinking   alcohol   was  18.4%;  Ta   Van   Binh  indicated the rate of smoking was 20.7% and drinking was 22.9%.

The rate of overweight, obesity was 50,1%; there was nodifference  between males and female (Table 3.7). Ta Van Binh indicated the rate was  4.3% and 52.0%;, respectively.

In people with pre­diabetes, percentage of abdominal obesity (50.1%) in  men (53.2%) was higher than women  48.5% (p<0.05)  (Figure 3.1).  Nguyen  Kim Cuc, Tran Huu Dang showed the rate was 47.2 of abdominal obesity.

Hypertention: Prevalence of hypertension (JNCVII)  was  36.1%;  pre­ hypertention   was  53.4%.   There   was   differencein   prevalence   of  hypertension between men and women. Nguyen Kim Cuc, Tran Huu Dang  showed the ratehypertension was 23.1%; Phan Long Nhon was 30.15% in  pre­diabetes in Binh Dinh province.

Dyshyperlipedia:   In   pre­diabetes:   Prevalence  of   dyslipidemia  was  89.8%  (Table  3.9);  mainly  Cholesterol,  LDL­C  và  Triglycerit  disorders.  Tran Thi Doan, Nguyen Vinh Quang indicated the rate was 86,9% in males  and 74.7% in females.

Metabolic   sydrome:  Prevalence   of   metabolic   syndrome  (ATPIII  criteria)  was  higher  than  IDF   criteria.  Prevalence  of  MS  in   female  was  higher than that in male (p<0.01). Nguyen Thi Thu Thao showed the rate of  69.2% (Table 3.10) 4.2.2. Insulin resistance, beta cell function, insulin sensitivity 4.2.2.1. Insulin resistance, beta cell function, insulin sensitivity

+ In pre­diabetes group, insulin index, HOMA2­IR index was higher;  conversely, the HOMA2­%S was lower than the control group; There was  no difference in HOMA2­% B index. The same results with the research of  Nguyen Duc Hoan.

+  There   wasn’t   difference   in   the   rate   of  pre­diabetes   with  insulin  resistance  between men and womenTable 3.12, Table 3.13 và Figure 3.4.  The same results with the research of Chang A.M. (2006).

+ The rate of insulin resistance, impaired beta cell function, impaired  insulin sensitivity  in  pre­diabetes  ranges  from  44.7%  to 71%.  Meye  C.,  Kanat M showed the same results in pre­diabetes.

4.2.2.2. The relationship, correlation between insulin resistance, beta cell   function, insulin sensitivity with risk factors

+ In pre­diabetes, there was a difference in HOMA2­%B between men  and women. There was no difference in HOMA2­% S, HOMA2­IR in the  pre­diabetes by gender (Figure 3.4, Table 3.13, Table 3.15).

+ In pre­diabetes, there was no correlation between insulin resistance,  beta   cell   function,   insulin   sensitivity   with   age   (p>   0.05)  Table   3.14.  Ferrannini E. 1996 indicated no correlation.

+   There   was   a   correlation   between   insulin   index,   HOMA2­IR,  HOMA2­%S with BMI index. There is no correlation between beta cell  function with BMI index Table 3.16.

+  There   was   no   relationship   between   insulin   resistance,   insulin

sensitivity, beta cell function with hypertension in pre­diabetes Table 3.17.

+   HOMA2­IR,   HOMA2­%B   index   in   pre­diabetes   group   with  abdominal   obesity   was   higher   than   in   non­abdominal   obesity   group.  Conversely, HOMA2­% S was lower than non­abdominal obesity group  (p<0.01) Table 3.18.

+ In pre­diabetes, there was a difference in insulin  index, HOMA2­ %B,   HOMA2­%S,   HOMA2­IR  between   group   with   and   without  dyslipiemia (Table 3.19).

+ In pre­diabetes with MS, insulin index was higher than in group without  MS;  there   was  no  difference   in   HOMA2­%B,   HOMA2­%S,   HOMA2­IR  between groups with and without MS (Table 3.20 và Table 3.21).

+ Different forms of pre­diabetes vary in insulin  index, HOMA2­IR,  HOMA2­%S, HOMA2­%B; Insulin resistance, impaired insulin sensitivity  increased  from IFG, IGT, and IFG & IGT. The rate of insulin resistance,  impaired   insulin   sensitivity,   impaired   beta   cell   function   decreased   from  IGT, IGT&IFG and the last was IFG. 2.1. The Results of intervention, treatment 2.5.1. Changes in risk factors after 24 months of intervention 2.5.1.1. Changes in habit

Interventions reduced  the rate of eating less vegetables and regularly  eat fat; However, the rate of smoking, alcohol  drink  changed very  little.  Cao My Phuong, Dinh Thanh Hue, Nguyen Hai Thuy indicated reduction  in the rate of smoking, drinking alcohol Figure 3.5.

The rate of smoking, drinking alcohol  without changes; however the  results showed that decrease in the rate of lack of eating fruits/vegetables,  especially in compliance group.

2.5.1.2. Changes in phisical activity

Vigorous  work   decreased;   The   average   rate  of  moderate  phisical  activities, cicling increased after 24 months of intervention Figure 3.6. This  result clearly showed the effectiveness of the intervention.

Vigorous  work   decreased;   The   average   rate  of  moderate  phisical  activities, cicling increased after 24 months of intervention Table 3.26. The  results showed no difference statistically in Weight, BMI, WC, HC index  after   intervention.   Kwame   Osei,   Rhinesmith   (2004)   indicated   Weight,  BMI, WC, HC increased after 6 years.

The prevalence of overweight and obesity of the compliance group

was significantly lower than that of the non­compliance group. 2.5.1.3. Changes in Lipid

The results of changes in Lipid showed in Table 3.27, Table 3.28 and  Figure 3.8. After intervention, The mean of TC, LDL­C index decreased  significantly;   Conversely,   the   mean   of   TG,   HDL­C   index   increased  significantly. There was no different in the mean of TC, TG, LDL­C index  between   compliance   and   non­compliance   group.   The   mean   of   HDL­C  index   in   the   non­compliance   group   was   significantly   lower   than   the  compliance group.

After   intervention,  the  prevalence  of   pre­diabetes  with   TC,  LDL­C  disorder and dyslipidemia decreased significantly; conversely, the rate of  TG, HDL­C disorder increased significantly. There was no difference in  the prevalence of lipid disorders between the compliance group and the  non­compliance group (Kwame Osei, Rhinesmith ­ 2004). 2.5.2. Changes in blood glucose after intervention

After   intervention,   fasting   glucose   index,   glucose   2h   decreased  significantly compared to baseline values. Fasting blood glucose, glucose  2h   in   the   compliance   group   decreased   and   lower   than   that   in   non­ compliance group.

IFG & IGT group had the highest rate of diabetes; IGT group had the  lowest rate of diabetes, after 24 months. The incidence of diabetes in the  non­compliance group was higher than the compliance group. William C  Knowler (2001), Daqing(1997) indicated the effectively reduction in blood  glucose by lifestyle intervention. 2.5.3. Changes   in   beta   cell   fucntion,   insulin   sensitivity,   insulin  resistance

Normal   glucose   group,   insulin   index,   insulin   resistance   in   normal  group was lowest; beta cell fuction and insulin sensitivity was higher than  those in pre­diabetes and diabetes group (p <0.01).

After intervention, the mean of insulin index, insulin resistance, beta­ cell   fuction   and   insulin   sensitivity   index   increased   significantly.   Insulin  index in compliance group didn’t increase; there was diference statistically  between   compliance   group   campare   with   non­compliance   groups   Table  3.34 và Table 3.35, Figure 3.10. In accordance with Ferrannini, Gordon C.  Weir. 's observations.

Table   3.34,   Table   3.37,   Figure   3.12.   showwed   the   results:   After  intervention,   the   incidence   of   hyperinsulinemia   in   the   non­compliance  group was significantly higher than the compliance group.

HOMA2­IR index in the compliance group was unchanged compared  with   the   group   of   non­compliance   increased   significantly,   p   <0.01.  HOMA2­IR   index   in   the   non­compliant   group   was   higher   than   the  compliance group. HOMA2­IR index (insulin resistance) in the compliance  group was lower significantly than the non­compliance group. The mean  HOMA2­%   S   (insulin   sensitivity)   index   in   the   compliance   group   was  higher than the non­compliance group.

After   intervention,   the   rate   of   decrease   in   insulin   sensitivity   in   the  compliance group was lower than the non­compliance group. HOMA2­% S  (insulin sensitivity) increased in the compliance group was higher than the non­ compliance group. HOMA2­%B index in both groups increased, and there was  no difference between the two groups after intervention (p> 0.05).

Table   3.34,   Table   3.37   showed   that:   After   24   months,   the   risk   of  developing diabetes in insulin resistance group was 1.3 times more (95%  CI 0.81­2.11). The risks of developing diabetes in the compliance group  was lower than that in non­compliance group.

CONCLUSION

1. Risk factors, beta cell fuction, insulin resistance

­ Age: of pre­diabetes mainly from 40 ­70s. ­ Habit:  Prevalence of  smoking was 33.2%, alcohol drinking 73.2%;  lack   of   eating   fruits/vegetables12.9%,   fat   using   12.1%.  Vigorous  and  moderate in work was 67.3%; in phisical activity 47.9%; traveling 60.8%.

­ Phisical   Index:The   rate   of   overweight,   obesity   was   50.1%;

Hypertention 36.1%, pre­hypertention 53.4%.

­ MS   and   Dyslipidemia:  Prevalence   of   MS   was   51.7%   (ATPIII),

29.7% (IDF); 88.8%with dylipidemia, maily TC, LDL­C disorders.

­ Insulin,   HOMA2:  insulin   index,   HOMA2­IR   index   was   higher;

conversely, the HOMA2­%S was lower than the control group.

­ The  relationship, correlation:There was a  correlation between  the  insulin,   insulin   sensitivity   and  beta   cell   function  with   BMI,   abdominal  obesity, MS and blood glucose status 2. The Results of intervention after 24 months

activities,

­ Risk  factors:  The   rate   of   smoking,   alcohol  drinking   changed  very  little; eating less vegetables and regularly eat fat reduced. The rate of doing  moderatephisical  cicling  increased.   The   results   showed  difference   statistically   between   the   compliance   group   and   the   non­ compliance group.

­ Glucose:  Fasting   glucose,   glucose   2h   decreased   significantly  compared to baseline values; of those in the compliance group decreased  and lower than that in non­compliance group. IFG & IGT group had the  highest rate of diabetes; the incidence of diabetes in the non­compliance  group was higher than the compliance group.

­ Insulin index, HOMA2:  Insulin index, insulin resistance, beta cell  fuction   and   insulin   sensitivity   index   increased   significantly;   there   was  diference   statistically   between   compliance   group   campare   with   non­ compliance groups

­ Risk   of   developing   diabetes:  the   risk   of   developing   diabetes   in  insulin resistance group was 1.3 times more (95% CI 0.81­2.11). The risks  of developing diabetes in the compliance group was lower than that in non­ compliance group.

RECOMMENDATIONS

1. The prevalence of diabetes among people over over 40 years of age is  very   high  should  to  be  screened  to   detect   diabetes   and  pre­diabetes   for  early intervention in time to reduce morbidity.

2. Interventions programme shows the effectiveness of diabetes prevention  should to be replicated in other provinces. Compliance indicates that will  reduce risk factors and the risk of devenloping diabetes. Insulin resistance,  insulin   sensitivity,   beta   cell   fucntion   can   be   used   to   predict   the   risk   of  developing diabetes in pre­diabetic subjects after 24 months.