Adherence to the DASH and Mediterranean diets is associated with decreased risk for gestational diabetes mellitus

Accepted Manuscript Adherence to the DASH and Mediterranean diets is associated with decreased risk of gestational diabetes mellitus Vajihe Izadi, MSc, Hotav Tehrani, MD, Fahimeh Haghighatdoost, PhD, Atefeh Dehghan, BSc, Pamela J. Surkan, ScD, Leila Azadbakht, PhD PII:

S0899-9007(16)00132-5

DOI:

10.1016/j.nut.2016.03.006

Reference:

NUT 9740

To appear in:

Nutrition

Received Date: 3 November 2015 Revised Date:

29 January 2016

Accepted Date: 7 March 2016

Please cite this article as: Izadi V, Tehrani H, Haghighatdoost F, Dehghan A, Surkan PJ, Azadbakht L, Adherence to the DASH and Mediterranean diets is associated with decreased risk of gestational diabetes mellitus, Nutrition (2016), doi: 10.1016/j.nut.2016.03.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

ACCEPTED MANUSCRIPT Adherence to the DASH and Mediterranean diets is associated with decreased risk of

1

gestational diabetes mellitus

2

Vajihe Izadi (MSc)1, 2, Hotav Tehrani (MD)3, Fahimeh Haghighatdoost (PhD)1,2, Atefeh Dehghan (BSc)1,2, Pamela J. Surkan (ScD)4, Leila Azadbakht (PhD)1,2,5

3 4 5 6

Food Security Research Center, Isfahan University of Medical Sciences, Isfahan, Iran.

2

Department of Community Nutrition, School of Nutrition and Food Science, Isfahan

RI PT

1

University of Medical Sciences, Isfahan, Iran.

Department of Obstetrics and Gyncology, Isfahan University of Medical Sciences, Isfahan,

SC

3

Iran

Department of International Health, John Hopkins Bloomberg School of Public Health, USA

M AN U

4

5

Department of community nutrition, School of Nutritional Sciences and dietetics, Tehran University of Medical Sciences, Tehran, Iran.

7 8 9 10 11

Corresponding author:

16

TE D

Running title: DASH, Mediterranean diets and GDM

12 13 14 15

17

Department of Community Nutrition,

18

School of Nutrition & Food Science

19

EP

Dr. Leila Azadbakht with below addresses:

Isfahan University of Medical Sciences, Isfahan, Iran

AC C

E-mail: [email protected]

20 21

Fax: 03136682509

22

Phone: 03137922719

23

PO Box: 81745

24

Acknowledgment: Funding for this study was granted by the Food Security Research

25

Center, School of Nutrition and Food Science, Isfahan University of Medical Sciences,

26

Isfahan, Iran

27

2

ACCEPTED MANUSCRIPT 28

Background: Few studies have examined the association between adherence to the DASH

29

(Dietary Approaches to Stop Hypertension) or Mediterranean (MED) diets and prevalence of

30

gestational diabetes mellitus (GDM). The aim of the present study was to evaluate the

31

associations the between DASH and MED diets with GDM.

RI PT

ABSTRACT

32

33

and without (n=260) GMD. An average of three 24-hour dietary records was used to assess

34

SC

Methods: In a case-control hospital-based study, we recruited pregnant women with (n=200)

35

MED scores were calculated using the Trichopoulou method. GDM was defined as fasting

36

M AN U

participants’ dietary intakes. DASH scores were calculated based on the Fung method and

37

pregnancy. The risk of GDM was assessed across tertiles of DASH and MED scores.

38

Results: DASH and MED diets were negatively related to fasting blood sugar (FBS) and

39

HBA1c and serum triglyceride concentrations. HDL-C was significantly higher for those in

40

TE D

glucose (FG)>95 mg/dl or 1-hour post prandial glucose >140 mg/dl for the first time in the

41

serum cholesterol level was lower in the third tertile of the MED diet but not for the DASH

42

diet. Participants who were in the highest tertile of the MED diet had 80% lower risk for

43

EP

the top tertile of the DASH but not MED diet in comparison with the lowest tertile. Total

44

DASH eating plan was associated with 71% reduced risk of GDM (P trend = 0.006) after

45

adjustment for potential confounders.

46

Conclusion: Adherence to DASH or Mediterranean diets are associated with decreased risk

47

of GDM

48

Keywords: Dietary Approaches to Stop Hypertension, Mediterranean diet, Glucose

49

tolerance, Gestational diabetes mellitus, pregnant women

50

AC C

GDM compared to those in the lowest tertile (P trend = 0.006). Greater adherence to the

51

3

ACCEPTED MANUSCRIPT 52

Gestational diabetes mellitus (GDM) is defined as carbohydrate intolerance first recognized

53

during pregnancy (1). It affects 1.0 to 14% of all pregnancies worldwide (2). The prevalence

54

of GDM is 4.7% in Iran (3). It is positively related to short- and long-term complications for

55

both mothers and their newborns (4), including increased risk of diabetes, hypertension and

56

cardiovascular diseases in later life (5). Several strategies including dietary counseling,

57

lifestyle modification, moderate hypocaloric diet, dietary intervention and glucose lowering

58

agents have been used to manage GDM and improve pregnancy outcomes (6-11). Although

59

SC

RI PT

INTRODUCTION

60

diets are considered cornerstone approaches to manage GDM (6, 7, 11-13), according to a

61

recent meta-analysis no strong conclusions can be drawn as to the best intervention for

62

managing GDM (11). Because of the interactions between foods and nutrients, it is optimal to

63

study dietary patterns than to study them separately (4). Also, due to differences in dietary

64

patterns between Western and Eastern societies, epidemiologic studies in both contexts

65

should be conducted.

TE D

M AN U

diet therapy focusing on fatty acids, micronutrient supplementation and low glycemic load

EP

The Dietary Approaches to Stop Hypertension (DASH) diet is a low-glycemic index and low-

66

67 68

potassium and dietary fiber (2). The DASH diet was initially suggested for hypertensive

69

AC C

energy dense dietary pattern, which contains high quantities of phytoestrogens, magnesium,

patients (5), but it is also an effective approach for improving cardiovascular risks, diabetes

70

and metabolic syndrome (2, 5). The Mediterranean diet (MED diet), emphasizing

71

consumption of fruits, vegetables, legumes, whole grains and foods rich in monounsaturated

72

fatty acids (MUFA), and is associated with lower risk of many chronic diseases (14, 15).

73

Most available research on this topic has been focused on western populations. However, due

74

to substantial differences in dietary patterns between middle-eastern and western populations,

75

it is useful to examine the overall effects of these dietary patterns in epidemiological studies

76

4

ACCEPTED MANUSCRIPT 77

consumption patterns of different components of the DASH and MED diets in Iranian

78

populations differ from those in other countries. Also, most research has examined this

79

association in women with a prior history of GDM (16, 18). Therefore, evaluating the

80

association between adherence to healthy dietary patterns, such as DASH and MED diets,

81

and incidence of GDM in Iranians might add to current knowledge. Our aim was to evaluate

82

the associations between DASH and MED dietary patterns with the prevalence of gestational

83

diabetes mellitus.

84

SC

RI PT

conducted in diverse settings (2, 4, 16, 17). This is of particular importance given that the

M AN U

Subjects and Methods:

85

86

Nutrition Clinic of Isfahan and Shahid Beheshti Hospital among 463 pregnant women

87

carrying singleton fetuses who were 22-44 years of age and between weeks 5-28 of

88

pregnancy; 263 of the women were healthy and 200 had GDM. For every pregnant women

89

TE D

Participants: This hospital-based case-control study was conducted in the Azzahra Hospital,

90

pregnancy was chosen for the control group. Pregnant women having abnormal fasting

91

glucose (FG), i.e. FG>95 mg/dl or 1-hour post prandial glucose>140 mg/dl for the first time

92

EP

with GDM selected in a specific week of the pregnancy, another woman in the same week of

93

had abnormal energy intake (<800 or >4200 kcal/day), Type 1 and 2 diabetes, cancer, and

94

cardiovascular diseases. Participants with incomplete daily food records and health

95

information were also excluded. Written informed consent was obtained from all participants.

96

The present study was approved by the Isfahan University of Medical Sciences, Isfahan, Iran.

97

Assessment of dietary intake: Participant dietary intake was estimated from the average of

98

three 24-hour dietary records. Participants were counseled by a trained dietitian regarding

99

how to complete dietary food record forms and they were checked by the dietition to ensure

100

AC C

in pregnancy were selected. We did not include women who were carrying multiple fetuses,

5

ACCEPTED MANUSCRIPT 101

them. We used household measures to convert all portion sizes to grams. Food records were

102

analyzed using Nutritionist IV software modified for Iranian foods (First Databank Division,

103

The Hearst Corporation, San Bruno, CA, USA).

104

RI PT

their accuracy. Incomplete forms were followed up with phone calls in order to complete

105

MED dietary scores (19). Accordingly, there were a maximum of 9 points, counting 1 point

106

if: the daily serving of fruits, fish, vegetables, whole grains, legumes, nuts and ratio of grams

107

of MUFA to saturated fatty acids (SFA) were equivalent to or greater than the median intake

108

of study population and also the daily serving of meats (red meat, poultry and processed

109

M AN U

SC

Creation of dietary scores: We followed the methodology of Trichopoulou et al. to calculate

110

adjustment, using the residual method, was done for all food groups before the score ranking.

111

Finally, we categorized participants according to tertiles based on their scores.

112

Calculating the DASH dietary score according to the Fung method (20), one point was

113

TE D

meats) and dairy products were less than the median intake of the study population. Energy

114

whole grains consumed were within the highest quintile of the study population. A point was

115

also given if daily intake of sodium, red and processed meats or sweetened beverages were in

116

EP

received if participant intake of fruits, vegetables, nuts and legumes, low-fat dairy products or

AC C

the lowest quintile of intake of the study population.

117

Assessment of anthropometric variables and blood pressure: Participant weight was

118

recorded to the nearest of 100 g using SECA scales, while wearing light clothing and no

119

shoes. Height was measured using tape, when not wearing shoes. Body mass index (BMI)

120

was calculated by weight (kg)/height (m2). Systolic and diastolic blood pressures (SBP &

121

DBP) were measured in duplicate after at least 5 minutes of rest in sitting position and being

122

calm. Appropriate cuff-sizes according to arm size and SBP were defined as a clearing of the

123

first sound of pulsation (first Korotkoff phase) and DBP was defined as disappearance of the

124

6

ACCEPTED MANUSCRIPT 125

recorded as the final value.

126

Assessment of other variables: Trained health care providers asked women about socio-

127

demographic information. The three socio-demographic categories based on the distribution

128

of responses were: high (scores>66%), middle (scores 33%-66%) and low (scores <33%).

129

Blood samples were centrifuged within 30-45 minutes of collection for 10 minutes at 500×g

130

and at 40C. Samples were analyzed using an auto-analyzer (Selectra 2; Vital Scientific,

131

Spankeren, Netherlands). ALT, AST, HDL-cholesterol, LDL-C, fasting glucose and serum

132

level of total cholesterol were measured using commercially availableenzymatic kits (Pars

133

M AN U

SC

RI PT

sound (fifth Korotkoff phase). The average of two blood pressure measurements was

Azmmoun, Tehran, Iran). Triglycerides were measured with glutathione oxidase.

134

Statistical analysis:

135

136

collected during their pregnancies. Higher scores indicated higher compliance and the lower

137

scores corresponded to lower compliance. We used one-way ANOVA for continuous

138

variables (e.g. BMI, age, blood pressure and biochemical markers). For the categorical

139

variables, we used X2 tests to assess the distribution of categorical variables across the tertiles

140

of DASH and MED scores. The Mantel-Haenszel test was used to calculate p-values for

141

AC C

EP

TE D

We categorized subjects based on tertiles of DASH and MED dietary patterns scores

trends across the tertiles of dietary patterns. Dietary intakes were compared using ANOVA.

142

We performed multivariable logistic regression to find the relation between either DASH or

143

MED diets and GDM risk. Theses associations were examined in both crude and adjusted

144

models. In logistic regression models, we used crude models without adjustment and two

145

adjusted models to control for the effects of potential confounders. We controlled the

146

confounding effects of age and energy in the first model and additionally controlled for the

147

number of children and socio-economic status (education, occupation, and economic status)

148

7

ACCEPTED MANUSCRIPT (21) in the second model.

Statistical analyses were performed by using the SPSS for

149 150

RESULTS

151

We found that cases had lower adherence to the DASH diet than the controls (22.73±4.06 vs.

152

25.03±4.92; P<0.0001), but observed no significant differences in adherence to the MED diet

153

(4.01±1.57 vs. 4.92±1.49; P=0.81). Additionally, serum levels of fasting blood sugar, HbA1c,

154

LDL-c and triglycerides were significantly higher in cases than controls. No differences were

155

observed between cases and controls for demographic variables. Characteristics of

156

SC

RI PT

Windows software (version 16.0). P < 0.05 was considered as statistically significant.

157

socio-economic status did not differ across tertiles of adherence to the MED and DASH diets.

158

We found that women in the highest tertiles of DASH and MED scores had significantly

159

lower levels of FBS, blood pressure and HbA1C (P<0.05). Serum levels of HDL-cholesterol

160

were substantially higher among participants with scores in the top tertile of DASH diet

161

scores in comparison to those in the lowest (48.35±9.22 vs. 46.40±9.83 mg/dl, P=0.004).

162

TE D

M AN U

participants across the tertiles of dietary patterns are shown in Table 1. Participant BMI and

163

adherence to the MED diet compared to those in the first tertile (166.39±34.60 vs.

164

179.76±34.51 mg/dl, P=0.03).

165

Maternal dietary intakes of participants across the tertiles of MED and DASH scores are

166

AC C

EP

Also, serum total cholesterol level was significantly lower in women with maximum

shown in Table 2. Pregnant women within the highest DASH and MED diet tertiles had

167

significantly higher carbohydrate, but lower fat and protein intakes compared those in the

168

first tertile (P=0.001). The daily intakes of SFA and cholesterol among adherents of MED

169

diet (in the third tertile) were significantly lower than those with poor adherence (11.54±3.69

170

vs. 15.96±8.69 g/d for SFA and 109.20±36.86 vs. 152.98±77.44 g/d for cholesterol,

171

P=0.001). Although the intake of SFA among pregnant women in the highest tertile of the

172

DASH diet was significantly greater than those in the lowest tertile (15.42±7.23 vs.

173

8

ACCEPTED MANUSCRIPT 174

(146.09±53.22 vs. 125.93±56.71 g/d, respectively, P=0.0001). Pregnant women in the highest

175

tertiles of the DASH and MED diets consumed more vegetables, fruits, legumes, fish, whole

176

grains, and dietary fiber (P<0.05).

177

RI PT

12.46±5.27 g/d, respectively, P=0.0001), daily intake of cholesterol was significantly lower

178

Table 3. Both dietary pattern adherence scores were significantly and inversely related to the

179

risk of GDM. In the crude model, for participants in the third tertile compared with those in

180

the first tertile, the risk of GDM was lower with the MED (OR: 0.22; 95% CI: 0.13, 0.37; P =

181

0.006) and DASH diets (OR: 0.32; 95% CI: 0.20, 0.52; P = 0.0001). Also, in the fully

182

M AN U

SC

Multivariable-adjusted odds ratios for GDM across tertiles of dietary patterns are shown in

183

tertile, had lower risk of GDM with both the MED (OR: 0.20; 95% CI: 0.50, 0.70; P = 0.006)

184

and DASH diets (OR: 0.29; 95% CI: 0.17, 0.48; P = 0.006). We found that adherence to

185

MED dietary pattern was more strongly associated with reduced risk of GDM (80%)

186

compared to adherence to the DASH diet (71%).

187

DISCUSSSION

188

TE D

adjusted multivariable model, participants in the third tertile compared with those in the first

EP

Our results indicate that adherence to the DASH and Mediterranean diets were associated

189 190

patterns to investigate correlations observed in the present study. This type of study

191

contributes to furthering our understanding of the associations between diet and diseases in

192

specific regions of the world, where social factors and dietary patterns may be distinct.

193

Epidemiologic studies in the field of nutrition are rare in Iran and longitudinal studies are

194

needed to confirm associations between dietary patterns and chronic diseases.

195

A previous randomized controlled trial showed that consumption of the DASH diet for four

196

weeks among women with GDM resulted in improved pregnancy outcomes (2). According to

197

AC C

with decreased risk of gestational diabetes. We analyzed a priori DASH and MED dietary

9

ACCEPTED MANUSCRIPT 198

lower risk of GDM, respectively (14). Also, adherence to DASH and MED diets was

199

associated with 46% and 40% lower risk of type 2 diabetes, respectively among women with

200

history of GDM (16). Another study reported that consumption of the DASH diet for four

201

weeks among pregnant women with GDM had favorable impacts on glucose tolerance in

202

comparison with a control group (17). In a three year-long prospective randomized trial, the

203

Mediterranean lifestyle intervention, which focused on diet and physical activity among

204

women with prior GDM, led to a lower glycemic response (42.8%) among those in the

205

SC

RI PT

a cohort study, adherence to the DASH and the MED diets was correlated with 34% and 24%

206

of 1,070 consecutive pregnant women, adherence to the MED diet was correlated with lower

207

incidence of GDM and higher degree of glucose tolerance (22).

208

Several mechanisms may explain the health benefits of DASH and MED dietary patterns.

209

High quantities of fruits and vegetables, less red and processed meat, emphasis on

210

carbohydrate quality and intake of unsaturated vegetable oils are common components in

211

these dietary patterns. High intakes of dietary fiber, magnesium, potassium, vitamin C and

212

phytochemical due to vegetables and fruits consumption in these diets might contribute to the

213

EP

TE D

M AN U

intervention group, compared with the control group (56.7%) (18). In one observational study

214

and vegetables are correlated to lower risk of metabolic syndrome (23). Consumption of

215

AC C

beneficial preventive effects on GDM (17, 2). We previously found that high intakes of fruits

whole grains with low GI and lower intake of simple sugars in these diets could decelerate

216

absorption of glucose and insulin requirements (24). Also we observed that whole grain

217

intake beneficially affects markers of systemic inflammation, which play an important role on

218

the incidence of diabetes (25). Consumption of PUFA and MUFA in vegetable oils in these

219

dietary patterns could improve glucose tolerance (17). Also, the low sodium content of these

220

healthy dietary patterns is known to be involved in the prevention of metabolic disorders

221

10

ACCEPTED MANUSCRIPT 222

improving glucose tolerance through multiple mechanisms (17).

223

Strengths of this study include an adequate sample size, allowing adjustment for several

224

potential confounders. The main limitation of the present study is its observational case-

225

RI PT

(26). Potassium, calcium and magnesium from the DASH diet also might be responsible for

226

a causal relationship. Our study was limited to only three 24-hours dietary records to assess

227

dietary intake, which may result in attenuation of our results. Also, residual confounding due

228

to incomplete adjustment for unknown confounders may have led to a change in or have

229

weakened our results.

230

M AN U

SC

control design. Larger prospective cohort studies would help to provide stronger evidence for

231

DASH or MED dietary patterns had lower risk of GDM. Future studies are needed in the

232

context of countries with non-Western diets to confirm these results and to investigate the

233

relation between dietary patterns and risk for GDM longitudinally.

234

AC C

EP

TE D

In conclusion, based on the present case-control study, pregnant women who followed the

235

11

ACCEPTED MANUSCRIPT 236

1. Maymone AC, Baillargeon JP, Ménard J, Ardilouze JL. Oral hypoglycemic agents for

237

gestational diabetes mellitus? Expert Opin Drug Saf.2011; 10:227-38.

238

2. Asemi Z, Samimi M, Tabassi Z, Esmaillzadeh A. The effect of DASH diet on pregnancy

239

outcomes in gestational diabetes: a randomized controlled clinical trial. Eur J ClinNutr,

240

RI PT

REFERENCES

2014;68:490-5.

241 242

pregnancy: a systematic literature review. J Nutr Metab 2010; 2010: 282464.

243

4. Tobias DK, Zhang C, Chavarro J, Bowers K, Rich-Edwards J, Rosner B, et al.

244

Prepregnancy adherence to dietary patterns and lower risk of gestational diabetes mellitus.

245

Am J Clin Nutr, 2012;96:289-95.

246

M AN U

SC

3. Louie JC, Brand-Miller JC, Markovic TP, Ross GP, Moses RG. Glycemic index and

247

clinical trial investigating the effect of DASH diet on insulin resistance, inflammation, and

248

oxidative stress in gestational diabetes. Nutrition, 2013; 29:619-24.

249

6. Thomas B, Ghebremeskel K, Lowy C, Crawford M, Offley-Shore B. Nutrient intake of

250

women with and without gestational diabetes with a specific focus on fatty acids. Nutrition.

251

2006; 22:230-6.

252

TE D

5. Asemi Z, Samimi M, Tabassi Z, Sabihi SS, Esmaillzadeh A. A randomized controlled

253

outcomes. Curr Opin Endocrinol Diabetes Obes.2007; 14:465-9.

254

8. McGowan CA, McAuliffe FM. The influence of maternal glycaemia and dietary glycaemic

255

index on pregnancy outcome in healthy mothers. Br J Nutr. 2010; 104:153-9.

256

9. Khattab S, Mohsen IA, AboulFoutouh I, Ashmawi HS, Mohsen MN, van Wely M, et al.

257

Can metformin reduce the incidence of gestational diabetes mellitus in pregnant women with

258

polycystic ovary syndrome? Prospective cohort study. Gynecol Endocrinol. 2011; 27:789-93.

259

10. Artal R, Catanzaro RB, Gavard JA, Mostello DJ, Friganza JC. A lifestyle intervention of

260

weight-gain restriction: diet and exercise in obese women with gestational diabetes mellitus.

261

Appl Physiol Nutr Metab. 2007; 32:596-601.

262

AC C

EP

7. Lim SS, Noakes M, Norman RJ. Dietary effects on fertility treatment and pregnancy

12

ACCEPTED MANUSCRIPT 263

preventing gestational diabetes mellitus: a systematic review and meta-analysis. J Womens

264

Health (Larchmt). 2011; 20:1551-63.

265

12. Moses RG, Barker M, Winter M, Petocz P, Brand-Miller JC. Can a low-glycemic index

266

diet reduce the need for insulin in gestational diabetes mellitus? A randomized trial. Diabetes

267

Care, 2009;32: 996-1000.

268

RI PT

11. Oostdam N, van Poppel MN, Wouters MG, van Mechelen W. Interventions for

269

randomized controlled trial investigating the effects of a low-glycemic index diet on

270

pregnancy outcomes in gestational diabetes mellitus. Diabetes Care. 2011; 34:2341-6.

271

SC

13. Louie JC, Markovic TP, Perera N, Foote D, Petocz P, Ross GP, Brand-Miller JC. A

272

Gaetano G, Iacoviello L; on behalf of the MOLI-SANI study Investigators. Adherence to the

273

traditional Mediterranean diet and mortality in subjects with diabetes. Prospective results

274

from the MOLI-SANI study. Eur J Prev Cardiol. 2015 Feb 3.

275

15. Tognon G, Lissner L, Sæbye D, Walker KZ, Heitmann BL. The Mediterranean diet in

276

relation to mortality and CVD: a Danish cohort study.Br J Nutr. 2014 Jan 14;111(1):151-9.

277

16. Tobias DK, Hu FB, Chavarro J, Rosner B, Mozaffarian D, Zhang C. Healthful dietary

278

patterns and type 2 diabetes mellitus risk among women with a history of gestational diabetes

279

mellitus. Arch Intern Med, 2012; 12;172:1566-72.

280

17. Asemi Z, Tabassi Z, Samimi M, Fahiminejad T, EsmaillzadehA. Favourable effects of the

281

Dietary Approaches to Stop Hypertension diet on glucose tolerance and lipid profiles in

282

gestational diabetes: a randomised clinical trial. Br J Nutr. 2013; 109:2024-30.

283

AC C

EP

TE D

M AN U

14. Bonaccio M, Di Castelnuovo A, Costanzo S, Persichillo M, De Curtis A, Donati MB, de

18. Pérez-Ferre N, Del Valle L, Torrejón MJ, Barca I, Calvo MI, Matía P, et al. Diabetes

284

mellitus and abnormal glucose tolerance development after gestational diabetes: A three-year,

285

prospective, randomized, clinical-based, Mediterranean lifestyle interventional study with

286

parallel groups. Clin Nutr. 2015;34(4):579-85.

287

19. Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean

288

diet and survival in a Greek population. N Engl J Med. 2003; 348:599–608.

289

13

ACCEPTED MANUSCRIPT 290

Comparison of 4 established DASH diet indexes: examining associations of index scores and

291

colorectal cancer. Am J Clin Nutr. 2013; 98:794-803.

292

21. Garmaroudi GHR, Moradi A. Socio-economic status in Iran: a study of measurement

293

index. Payesh 2010; 9: 137-44. [In Persian].

294

RI PT

20. Miller PE, Cross AJ, Subar AF, Krebs-Smith SM, Park Y, Powell-Wiley T, et al.

22. Karamanos B, Thanopoulou A, Anastasiou E, Assaad-Khalil S, Albache N, Bachaoui M,

295

et al. Relation of the Mediterranean diet with the incidence of gestational diabetes. Eur J

296

ClinNutr. 2014; 68:8-13.

297

SC

23. Esmaillzadeh A, Kimiagar M, Mehrabi Y, Azadbakht L, Hu FB, Willett WC. Fruit and

298 299

841489-97.

300

M AN U

vegetable intakes, C-reactive protein, and the metabolic syndrome. Am J ClinNutr. 2006;

24. de Munter JS, Hu FB, Spiegelman D, Franz M, van Dam RM. Whole grain, bran, and

301

germ intake and risk of type 2 diabetes: a prospective cohort study and systematic review.

302

PLoS Med 2007;4:e261.

303 304

intake favorably affects markers of systemic inflammation in obese children: a randomized

305

controlled crossover clinical trial. Mol Nutr Food Res. 2014; 58:1301-8.

306

TE D

25. Hajihashemi P, Azadbakht L, Hashemipor M, Kelishadi R, Esmaillzadeh A. Whole-grain

307

controversy resolved? Curr Hypertens Rep. 2012; 14, 193–201.

308

AC C

EP

26. Alderman MH & Cohen HW. Dietary sodium intake and cardiovascular mortality:

ACCEPTED MANUSCRIPT

14 Table 1. Characteristics of adherence among tertiles of dietary patterns1

Tertile 3 160 26.51±5.35 29.21±4.30

70.70 5.40 0.70 22.40 0.70

69.90 4.10 3.40 21.20 1.40

82.90 12.70 4.40

92.10 4.60 3.30

85.20 6.70 8.10

90.50 0.60 3.80 5.10

96.70 0.00 0.70 2.60

95.30 0.00 0.70 4.00

50.60 25.30 24.10 51.90 36.10 12.00 1.30

44.10 36.20 19.70 61.20 28.30 10.50 0.00

M AN U

72.80 3.20 0.60 17.70 5.70

Ptrend 0.0001 0.29 0.05

Tertile 1 160 27.96±6.28 30.16±4.89

Med Diet Tertile 2 Tertile 3 134 165 29.85±6.77 27.18±5.27 29.64±5.05 29.40±4.73

RI PT

Tertile 1 134 28.65±6.99 29.94±5.09

72.70 3.80 1.50 18.20 3.80

65.80 5.00 0.60 25.50 3.10

75.30 3.80 2.50 17.10 1.30

82.80 11.20 6.00

86.70 12.10 1.20

90.00 1.20 8.80

91.00 0.70 3.00 5.20

94.50 0.00 1.20 4.20

96.20 0.00 1.20 2.50

47.00 26.90 26.10 56.70 32.10 11.20 1.50

47.30 30.90 21.80 52.10 37.00 10.90 0.00

46.90 29.40 23.80 57.50 33.80 8.80 0.00

44.00 23.90 9.70

46.10 23.60 10.90

77.50 15.60 1.20

0.02

46.20 29.10 8.20

TE D

EP 51.30 22.40 7.90

46.30 26.20 27.50 53.00 38.90 8.10 0.00

72.50 10.70 5.40

0.45

0.16

0.25

0.14 0.001

Ptrend 0.0001 0.38 0.39

<0.001

0.15

AC C

Number of subjects Age (year) BMI (kg/m2) Family history of (%) No disease Kidney Liver & kidney Cardiovascular Kidney and Cardiovascular Education (%) <12year 12 year >12 year Occupation (%) Housewife Office/employed Self-employed Retired Socio-economic status (%) poor moderate rich Number of children (%) 1-2 >2 Smoker (%) Use medications (%) Non users Lowering serum glucose agents Lowering glucose and hypertension

DASH Tertile 2 165 29.90±5.76 29.95±4.73

SC

Variables

0.91

0.81

0.08 <0.001

ACCEPTED MANUSCRIPT

15

7.40 14.20 12.10 2.50 4.00 8.20 7.30 3.10 96.08±36.75 0.001 118.82±54.35 112.81±46.70 95.21±40.77 12.44±1.72 0.002 12.46±1.75 11.80±1.55 12.45±1.70 7.55±0.81 0.14 7.64±0.90 7.44±0.88 7.61±.84 129.40±68.14 0.0001 161.50±67.44 152.19±76.29 131.65±55.90 48.35±9.22 0.004 45.66±9.30 46.06±5.30 47.65±8.92 92.37±35.21 0.28 98.35±28.21 94.71±31.84 91.17±36.05 167.34±35.28 0.09 179.76±34.51 178.23±69.52 166.39±34.60 21.71±15.40 0.60 20.47±13.71 22.21±12.30 19.89±11.66 23.20±8.08 0.14 21.37±8.43 23.81±21.17 22.30±7.56 6.04±1.59 0.0001 7.03±2.03 6.65±1.78 6.16±1.46

RI PT

9.90 8.60 114.71±52.06 11.83±1.52 7.46±.092 159.78±71.32 47.00±3.83 93.24±30.62 176.67±66.83 20.73±12.00 20.75±7.85 6.86±1.90

SC

10.80 5.70 114.05±51.58 12.38±1.75 7.66±0.89 153.47±69.76 46.40±9.83 97.84±31.39 179.31±42.78 20.29±9.74 23.72±21.52 6.75±1.75

M AN U

Lowering glucose and lipids Total of three FBS (mg/dl) SBP (mmHg) DBP (mmHg) TG (mg/dl) HDL (mg/dl) LDL (mg/dl) Total Cholesterol (mg/dl) ALT (mg/dl) AST (mg/dl) HbA1C (%) 1 Data presented are Mean±SD

0.0001 0.0001 0.08 0.0001 0.14 0.16 0.03 0.22 0.32 0.0001

2 P values presented resulted from analysis of variance and chi square tests

AC C

EP

TE D

3 BMI: body mass index; FBS: fasting blood glucose; SBP & DBP: systolic and diastolic blood pressure; TG: triglyceride; HDL & LDL: high and low density lipoprotein; ALT: alanine trance aminase; AST: aspartate trance aminase.

ACCEPTED MANUSCRIPT

16 Table 2. Dietary intakes of participants across tertiles of dietary patterns 1

Tertile 1 1932.84±35.30

Tertile 2 1791.21±36.35

Ptre nd Tertile 3 2171.86±36.89

0.0001

67.96±6.98

68.38±7.25

0.09

12.88±2.58

14.30±3.17

13.22±3.71

0.003

22.28±8.49

20.84±6.22

19.99±5.29

12.46±5.27

10.72±4.13

15.42±7.23

Cholesterol (g/d) Whole grains (g/d) Vegetables(g/d)

125.93±56.71

100.38±57.11

146.09±53.22

8.43±10.95

17.53±22.69

38.62±22.39

399.04±137.21

473.38±171.88

663.64±267.66

Low fat dairy (g/d) Meat (g/d)

357.62±194.71

390.31±176.11

59.81±26.05

62.35±43.86

Nuts (g/d)

8.40±7.79

14.26±17.72

Fruits (g/d)

545.21±270.27

623.50±233.81

Legumes (g/d)

16.06±6.45

12.84±6.16

Poultry (g/d)

9.68±5.01

Fish (g/d)

3.04±5.59

Sodium (mg/d)

Tertile 1 2126.07±39.36

Tertile 2 1846.19±35.82

Ptre nd Tertile 3 1948.40±36.23

0.0001

66.69±5.95

66.00±9.06

70.40±6.33

0.0001

14.27±3.20

12.80±3.31

0.001

13.26±2.96

M AN U

66.91±8.38

Med Diet

RI PT

Energy (kcal/day) Carbohydrate (% of total energy) Protein (% of total energy) Fat (% of total energy) SFA (g/d)

DASH

SC

Variables

21.59±5.92

22.77±8.88

18.85±4.20

0.0001

0.0001

15.96±8.69

11.58±3.77

11.54±3.69

0.0001

0.0001

152.98±77.44

114.85±49.37

109.20±36.86

0.0001

0.0001

12.48±20.14

16.86±21.28

33.11±22.36

0.0001

0.0001

438.31±179.14

506.60±277.79

572.27±185.16

0.0001

556.45±210.72

0.0001

519.78±225.94

400.39±232.70

393.92±149.50

0.0001

103.26±37.54

0.0001

74.79±44.20

61.85±32.78

88.03±42.97

0.0001

17.10±25.69

0.01

18.30±27.36

9.98±13.46

12.14±12.80

0.02

779.099±160.32

0.0001

578.89±226.87

591.53±282.51

772.49±179.91

0.0001

30.37±17.27

0.0001

14.69±5.83

17.32±12.25

25.19±16.94

0.0001

10.68±10.72

11.67±7.35

0.09

11.15±10.65

9.53±6.41

11.40±6.89

0.19

4.03±4.28

9.86±6.85

0.0001

3.09±3.57

3.87±5.36

9.43±7.37

0.0001

6656.30±2666.13 5398.18±3154.21 6408.69±3594.01

0.01

5758.69±2318.88 5790.72±3292.79 6874.87±4553.50 0.001

Iron (mgr/day)

12.09±2.93

11.51±3.00

14.01±3.75

0.05

13.18±3.23

12.30±3.80

12.20±3.07

0.001

Magnesium (mg/d)

342.85±101.57

347.79±86.81

422.21±97.32

0.0001

392.45±103.02

361.00±127.15

361.19±61.48

0.02

AC C

EP

TE D

0.01

ACCEPTED MANUSCRIPT

17

8.36±2.39

8.18±2.09

10.58±2.04

0.0001

10.05±2.51

8.49±2.66

8.70±1.79

0.001

Vitamin C (mg/d) Potassium (mg/d) Calcium (mg/d)

234.31±80.20

275.40±75.96

359.66±91.73

0.0001

246.51±64.31

273.66±115.89

339.29±76.62

0.0001

3971.13±1307.31 4050.02±1002.11 5109.69±1002.11

0.0001

4513.88±1242.02 4136.83±1421.39 4480.94±839.59

0.001

1110.50±395.92

1146.32±351.48

1535.62±600.80

0.0001

1429.10±425.21

1187.27±653.27

1194.42±297.51

0.02

B12 (mcg/d)

4.62±5.29

3.92±9.57

5.46±3.49

0.77

7.41±10.80

3.35±2.91

3.71±3.42

0.0001

26.08±7.93

29.03±7.90

0.004

26.39±9.41

26.47±10.81

26.73±8.62

0.01

SC M AN U

Dietary fiber 24.62±9.36 (g/d) 1 Data presented are mean±SD

RI PT

Zinc (mg/d)

AC C

EP

TE D

2 P-values presented are from analysis of variance

ACCEPTED MANUSCRIPT

18

Table 3. Multivariable-adjusted odds ratio and 95% confidence intervals for gestational diabetes mellitus (GDM) across tertiles of dietary pattern adherence scores in pregnancy. Tertile2

Tertile3

Ptrend

Model 1 Model 2 Model 3 DASH Diet

1 1 1

0.92 (0.58-1.47) 0.88 (0.54-1.42) 1.10 (0.67-1.80)

0.22 (0.13-0.37) 0.23 (0.14-0.39) 0.20 (0.50-0.70)

Model 1 Model 2 Model 3

1 1 1

0.81 (0.52-1.27) 0.83 (0.52-1.30) 0.90 (0.56-1.45)

0.32 (0.20-0.52) 0.30 (0.18-0.50) 0.29 (0.17-0.48)

M AN U

0.0001 0.0001 0.006

EP

TE D

Model 1 is the crude model. Model 2 is adjusted for age and energy. Model 3 is adjusted for age, energy, number of children and socio-economic status.

AC C

1

0.0001 0.0001 0.006

SC

Med Diet

RI PT

Tertile1

ACCEPTED MANUSCRIPT

Highlights Greater adherences to DASH and MED diets were associated with reduced risk of GDM. Participants in higher tertiles of DASH and MED diets had lower FBS and HbA1C.

AC C

EP

TE D

M AN U

SC

RI PT

Greater adherences to DASH and MED diets were associated with better lipid profiles.