The association of maternal ABO blood group with gestational diabetes mellitus in Japanese pregnant women

Accepted Manuscript Title: The association of maternal ABO blood group with gestational diabetes mellitus in Japanese pregnant women Authors: Masanori Shimodaira MD. PhD Teruyuki Yamasaki MD, PhD Tomohiro Nakayama MD, PhD PII: DOI: Reference:

S1871-4021(16)30025-X http://dx.doi.org/doi:10.1016/j.dsx.2016.03.003 DSX 576

To appear in:

Diabetes & Metabolic Syndrome: Clinical Research & Reviews

Received date: Accepted date:

20-2-2016 5-3-2016

Please cite this article as: Shimodaira M, Yamasaki T, Nakayama T, The association of maternal ABO blood group with gestational diabetes mellitus in Japanese pregnant women, Diabetes and Metabolic Syndrome: Clinical Research and Reviews (2016), http://dx.doi.org/10.1016/j.dsx.2016.03.003 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.

Original Article

The association of maternal ABO blood group with

ip t

gestational diabetes mellitus in Japanese pregnant women

cr

Subtitle: ABO blood group and GDM

an

us

Keywords: ABO blood group, diabetes, gestational diabetes, pregnancy, risk factor

d

Nakayama, MD, PhD2

M

Masanori Shimodaira, MD. PhD1,2, Teruyuki Yamasaki, MD, PhD3, Tomohiro

Department of Internal Medicine, Iida Municipal Hospital, Nagano, Japan

2)

Division of Laboratory Medicine , Department of Pathology and Microbiology, Nihon

Ac ce pt e

1)

University School of Medicine, Tokyo, Japan 3)

Department of Obstetrics and Gynecology, Iida Municipal Hospital, Nagano, Japan

Corresponding author: Masanori Shimodaira, M.D. Ph.D. Department of Internal Medicine, Iida Municipal Hospital, Nagano, Japan 438 Yawata-machi, Iida, Nagano, 395-8502, Japan Tel: +81 265-21-1255 (Ext. 5023); Fax: +81 265-21-1266 E-mail: [email protected] 1 Page 1 of 18

Abstract Aims: To investigated the association between the ABO blood group and gestational

ip t

diabetes mellitus (GDM). Materials and Methods: A retrospective case-control study

cr

was conducted using data from 5424 Japanese pregnancies. GDM screening was

us

performed in the first trimester using a casual blood glucose test and in the second trimester using a 50-g glucose challenge test. If the screening was positive, a 75-g oral

an

glucose tolerance test was performed for a GDM diagnosis, which was defined

M

according to the International Association of Diabetes and Pregnancy Study Groups. Logistic regression was used to obtain the odds ratio (OR) and 95% confidence interval

Ac ce pt e

d

(CI) adjusted for traditional risk factors. Results: Wwomen with the A blood group (adjusted OR: 0.34, 95% CI: 0.19 0.63), B (adjusted OR: 0.35, 95% CI: 0.18 0.68), or O (adjusted OR: 0.39, 95% CI: 0.21 0.74) were at decreased risk of GDM compared with those with group AB. Women with the AB group were associated with increased risk of GDM as compared with those with A, B, or O (adjusted OR: 2.73, 95% CI: 1.64 4.57). Conclusion: ABO blood groups are associated with GDM, and group AB was a risk factor for GDM in Japanese population.

2 Page 2 of 18

Introduction The ABO blood group system is a classification of blood types based on the presence or

ip t

absence of A and B antigens on red blood cells; it is the most generally used blood

cr

group classification in clinical practice. Recently, some studies have reported

us

associations between particular ABO phenotypes and an increased susceptibility to certain diseases including infection [1], cancer [2], vascular disease[3], and dementia

an

[4].

M

A few studies have examined the possible associations between ABO blood groups and the risk of type 2 diabetes mellitus (T2DM). Two studies, one from Malaysia

Ac ce pt e

d

[5] a and the other from India [6] have reported higher frequencies of the B group among patients with T2DM. On the other hand, in Nigerian subjects, the O (Rh negative) and A groups (Rh positive) were significantly higher among diabetic subjects than controls [7]. A prospective cohort study consisting of 82,104 women reported that groups A and B were at increased risk of T2DM compared with the O group [8]. Although some studies showed a link between ABO blood group and T2DM, the associations were not observed in others [9,10]. Gestational diabetes mellitus (GDM) is defined as glucose intolerance, which initiates or is diagnosed during pregnancy. This disorder is a common medical problem 3 Page 3 of 18

presenting increased risks of complications for both mother and child during pregnancy. In contrast with T2DM, literatures concerning the association of ABO blood types with

ip t

GDM are scarce. In Turkish subjects, Karagoz et al. reported a higher risk of GDM in

cr

the AB group patients [11]. However, this finding is in contrast with a Chinese study

us

that reported AB group was less likely to correlate with GDM [12]. Furthermore, a retrospective cohort study conducted in Thailand showed that the ABO blood group was

an

correlated with preeclampsia, but not with GDM [13]. The discrepancy in the results

M

from these studies may be partly due the variation in the genetic profiles among different ethnic groups. The aim of the present study was to determine whether the ABO

Ac ce pt e

d

blood group has an association with GDM in the Japanese population.

Materials and Methods Study population

Between October 2010 and September 2015, 5864 pregnancy women delivered at the Obstetrics Department of Iida Municipal Hospital, Nagano, Japan. In the present retrospective study, we collected information from the birth register. Cases with known type 1 or type 2 diabetes, taking medications affecting glucose metabolism, missing information on maternal height or weight, random glucose levels, and ABO blood 4 Page 4 of 18

groups were excluded. Women who had a fetus with a chromosomal or congenital abnormality were also excluded. To eliminate any possible bias arising from including

ip t

the same woman more than once, we used the date of most recent delivery if a woman

cr

had delivered twice or more in the hospital.

us

GDM screening and definitions

Based on the clinical recommendation by the Japan Society of Obstetrics and

an

Gynecology [14], all women underwent a universal two-step screening for GDM. In the

M

first trimester the casual blood glucose test with a cutoff value of 100 mg/dL was used, and in the second trimester a non-fasting 50-g blood glucose challenge test with a cutoff

Ac ce pt e

d

value of 140 mg/dL was used. If the screening was positive, a 75-g oral glucose tolerance test (OGTT) after an overnight fast was performed for the definite diagnosis of GDM. Overweight or obese women were recommended to undergo a 75-g OGTT at any time during gestation. GDM was defined according to the International Association of Diabetes and Pregnancy Study Groups [15] and the American Diabetes Association criteria [16] at a plasma glucose concentration equal or higher than one of the three following cut off points: 1) baseline, 92 mg/dL; 2) 1-h post OGTT, 180 mg/dL; or 3) 2-h post OGTT, 153 mg/dL. Informed consent was obtained from all participants. Informed consent was obtained from all participants. 5 Page 5 of 18

Pre-gestational weight was self-reported at the first prenatal visit. Information on behavioral factors and familial history of diabetes were collected through interviews.

ip t

Habitual smokers were subjects who continuously smoked one or more cigarettes per

cr

day for at least six months before pregnancy. Habitual drinkers were those who

us

consumed liquor on four or more occasions per month before pregnancy. Subjects consuming anti-hypertensive agents before pregnancy, or having a blood pressure of

an

>140/90 mmHg on more than two occasions before 20 weeks of gestation, were defined

M

as those with chronic hypertension [17]. Subjects with a familial history of diabetes had

Ac ce pt e

Laboratory analysis

d

one or more first degree diabetic relatives.

Plasma glucose samples collected in fluoride oxalate tubes were assessed within 2 h of sampling using the hexokinase method (GLU-HK; Shinotest Inc., Kanagawa, Japan). The ABO blood group was evaluated via agglutination technology using the Ortho® BioVue system (Ortho Clinical Diagnostics Japan, Tokyo, Japan) in the hospital. Statistical analysis

To compare the difference between the GDM and non-GDM groups, the Student’s t-test was used for continuous variables, and chi-squared test or Fisher’s test was used to compare the categorical parameters. Binary logistic regression analysis was performed 6 Page 6 of 18

to obtain the odds ratio (OR) and 95% confidence interval (CI) of ABO groups for GDM using both univariable and multivariable analysis. In multivariable analysis,

ip t

traditional risk factors for GDM including maternal age, pre-pregnancy body mass

cr

index (BMI), parity, chronic hypertension, non-singleton pregnancy, and familial history.

us

Data are expressed as means ± standard deviations. P values <0.05 were considered statistically significant. Statistical analyses were performed using the SPSS software

M

Results

an

version 21.0 (SPSS Inc. IL, USA).

Ac ce pt e

d

Of the 5484 women who visited Iida Municipal Hospital for delivery between January 2010 and September 2015, 5424 were included in the present study. Of the women included in analyses, 149 (2.75%) were diagnosed with GDM (Table 1). Compared with the non-GDM group, maternal age and pre-pregnancy BMI, frequency of multiparity, chronic hypertension, and familial history of diabetes were significantly higher among GDM group. Among the 5275 non-GDM subjects, the A, B, O, and AB blood groups were 1999 (37.9%), 1250 (23.7%), 1507 (28.6 %), and 519 (9.8 %), respectively, which were similar to those frequencies reported previously for the Japanese general population (A, 38.7%; B, 22.2%; O, 29.3%; and AB, 10.0%) [18]. The distribution of 7 Page 7 of 18

blood groups was significantly different between the non-GDM and GDM subjects; women with GDM were more likely to have the AB blood group.

ip t

The clinical and biochemical characteristics of subjects according to the ABO

cr

blood group are shown in Table 2. Maternal age, pre-pregnancy BMI, random glucose

us

level, frequency of multiparity, singleton pregnancy, chronic hypertension, and familial history of diabetes were similar among the four blood groups. In the univariable

an

analysis, women with A, B, and O blood groups were at a decreased risk for GDM; their

M

ORs versus the AB blood group were 0.42 (95% CI 0.26 0.66), 0.52 (95% CI 0.32 0.83), and 0.49 (95% CI 0.31 0.78), respectively (Table 3, Model M0).

Ac ce pt e

d

Compared with the non-AB group, the AB group had a significant increased risk of GDM (OR 2.23, 95% CI 1.46 3.36). Furthermore, similar results were obtained from adjusted models for covariates (Table 3, Model M1, M2, and M3); AB blood group had a significantly increased risk of GDM compared with other blood groups. Discussion

Several epidemiological and genetic studies concerning the association of T2DM and ABO blood groups are available in the literature. Because the ABO blood group phenotypes are stable throughout one's life time, it is thought to be important to identify the association between the phenotypes and the risk of glucose intolerance. GDM and 8 Page 8 of 18

T2DM share some characteristics including risk factors, pathophysiological mechanisms, and genetic susceptibility [19]. Recently, a few studies investigated the

ip t

possible association between ABO blood groups and risk of GDM. However, the results

cr

have been variable, inconsistent, and differed from one region to other [11-13].

us

A study on pregnant Iranian women showed that women with the AB blood group had higher fasting blood glucose levels in the second trimester than those with the

an

A blood group [20]. Karagoz et al. reported that the AB blood group seemed to be a risk

M

factor for GDM in a retrospective examination of 233 Turkish patients with GDM [11]. However, this study lacked information about traditional risk factors such as

Ac ce pt e

d

pre-pregnancy weight, BMI, and a family history of diabetes. In the present study, we aimed to investigate the relationship between ABO blood groups and GDM using a large cohort of Japanese population. The results indicated that the AB blood group had a risk for GDM compared with other blood groups after adjusting for confounders. The frequency of the Rh negative phenotype varies widely in different parts of the world; about 15% of Caucasian people are Rh negative, whereas only 0.5% of the Japanese population has the Rh negative blood type [21]. Because there were no subjects with the Rh negative blood type among the GDM group, we did not examine the association of the Rh blood type with risk of GDM. 9 Page 9 of 18

Our findings were inconsistent with a previously published study by Zhang et al. in which it was demonstrated using logistic regression analysis for adjustment of

ip t

traditional risk factors that the AB blood group was a protective factor against GDM in

cr

pregnant Chinese women [12]. In this study, the major blood group of non-GDM

us

subjects was B, accounting for 33.4% of the study participants. On the other hand, in the present study, the major group observed in the non-GDM subjects was A, accounting for

an

37.9% of the study participants. Even though both Japan and China are located in East

M

Asia, the distribution of ABO groups significantly varies between the races, ethnicities, and socioeconomic groups [22]. Therefore, our results indicate that the association

Ac ce pt e

d

between ABO blood group and risk of GDM may differ by population. Direct biological mechanisms underlying the association between ABO blood

groups and GDM are inconclusive. A recent genome-wide association study reported that a polymorphism at the ABO locus had a strong association with serum levels of tumor necrosis factor alpha (TNF-α) [23]. During pregnancy, TNF-α produced from the placenta plays an important role in insulin resistance pathways. Furthermore, ABO is a major locus for serum soluble E-selectin levels that have been associated with T2DM [24]. These genetic variants at the ABO locus may affect plasma inflammatory molecules and GDM risk. 10 Page 10 of 18

The present study had several limitations. First, because of a single-center cohort study, generalizations are probably limited. However, the distributions of blood

ip t

types in the non-GDM subjects are remarkably similar to those in the general Japanese

cr

population [18], indicating that the extent of the potential selection bias is rather

us

minimal. Second, there is a possibility of inclusion of GDM cases in the non-GDM group because not all subjects underwent the 75g-OGTT. The discrepancy may decrease

an

the statistical significance of the association between ABO blood groups and GDM.

M

Third, we were unable to assess the effects of the ABO blood type on plasma biomarker levels (such as TNF-α or soluble E-selectin) because of the unavailability of sufficient

Ac ce pt e

d

banked plasma. Furthermore, only limited variables could be included in the multivariate analysis. Other variables such as maternal physical activity, socioeconomic factors, or weight gain during pregnancy may have roles in GDM development, but these data are not currently available. Further epidemiological and genetic studies are necessary to define the relationship between ABO blood groups and GDM. In summary, among the Japanese general population, we found that pregnant

women with the AB blood group have a risk for GDM.

11 Page 11 of 18

Acknowledgements The authors thank Nobuo Shimosawa for preparing electronic database of subjects’

cr

ip t

medical records.

Declaration of interest statement

Ac ce pt e

d

M

an

us

The authors report no conflicts of interest.

12 Page 12 of 18

Reference

6. 7.

8.

9. 10.

11.

12.

ip t

cr

us

5.

an

4.

M

3.

d

2.

Cooling L: Blood Groups in Infection and Host Susceptibility. Clin Microbiol Rev 2015; 28: 801-870. Zhang BL, He N, Huang YB, Song FJ, Chen KX: ABO blood groups and risk of cancer: a systematic review and meta-analysis. Asian Pac J Cancer Prev 2014; 15: 4643-4650. Wu O, Bayoumi N, Vickers MA, Clark P: ABO(H) blood groups and vascular disease: a systematic review and meta-analysis. J Thromb Haemost 2008; 6: 62-69. Vasan SK, Rostgaard K, Ullum H, Melbye M, Hjalgrim H, Edgren G: ABO Blood Group and Dementia Risk--A Scandinavian Record-Linkage Study. PLoS ONE 2015; 10: e0129115. Kamil M, Al-Jamal HA, Yusoff NM: Association of ABO blood groups with diabetes mellitus. Libyan J Med 2010; 5: Sidhu LS, Malhotra P, Singh SP: ABO and Rh blood groups in diabetes mellitus. Anthropol Anz 1988; 46: 269-275. Okon UA, Antai AB, Osim EE, Ita SO: The relative incidence of diabetes mellitus in ABO/Rhesus blood groups in South-Eastern Nigeria. Niger J Physiol Sci 2008; 23: 1-3. Fagherazzi G, Gusto G, Clavel-Chapelon F, Balkau B, Bonnet F: ABO and Rhesus blood groups and risk of type 2 diabetes: evidence from the large E3N cohort study. Diabetologia 2015; 58: 519-522. Nemesure B, Wu SY, Hennis A, Leske MC: Hypertension, type 2 diabetes, and blood groups in a population of African ancestry. Ethn Dis 2006; 16: 822-829. Iyengar S, Hamman RF, Marshall JA, Baxter J, Majumder PP, Ferrell RE: Genetic studies of type 2 (non-insulin-dependent) diabetes mellitus: lack of association with seven genetic markers. Diabetologia 1989; 32: 690-693. Karagoz H, Erden A, Ozer O, Esmeray K, Cetinkaya A, Avci D, Karahan S, Basak M, Bulut K, Mutlu H, Simsek Y: The role of blood groups in the development of diabetes mellitus after gestational diabetes mellitus. Ther Clin Risk Manag 2015; 11: 1613-1617. Zhang C, Li Y, Wang L, Sun S, Liu G, Leng J, Guo J, Lv L, Li W, Hu G, Yu Z, Yang X: Blood group AB is protective factor for gestational diabetes mellitus: a prospective population-based study in Tianjin, China. Diabetes Metab Res Rev 2015; 31: 627-637.

Ac ce pt e

1.

13 Page 13 of 18

13.

18. 19. 20.

21.

22.

23.

us

an

M

17.

d

16.

Ac ce pt e

15.

cr

ip t

14.

Phaloprakarn C, Tangjitgamol S: Maternal ABO blood group and adverse pregnancy outcomes. J Perinatol 2013; 33: 107-111. Minakami H, Hiramatsu Y, Koresawa M, Fujii T, Hamada H, Iitsuka Y, Ikeda T, Ishikawa H, Ishimoto H, Itoh H, Kanayama N, Kasuga Y, Kawabata M, Konishi I, Matsubara S, Matsuda H, Murakoshi T, Ohkuchi A, Okai T, Saito S, Sakai M, Satoh S, Sekizawa A, Suzuki M, Takahashi T, Tokunaga A, Tsukahara Y, Yoshikawa H: Guidelines for obstetrical practice in Japan: Japan Society of Obstetrics and Gynecology (JSOG) and Japan Association of Obstetricians and Gynecologists (JAOG) 2011 edition. J Obstet Gynaecol Res 2011; 37: 1174-1197. Metzger BE, Gabbe SG, Persson B, Buchanan TA, Catalano PA, Damm P, Dyer AR, Leiva A, Hod M, Kitzmiler JL, Lowe LP, McIntyre HD, Oats JJ, Omori Y, Schmidt MI: International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy. Diabetes Care 2010; 33: 676-682. [Anonymous]: Diagnosis and classification of diabetes mellitus. Diabetes Care 2008; 31 Suppl 1: S55-60. [Anonymous]: ACOG Practice Bulletin No. 125: Chronic hypertension in pregnancy. Obstet Gynecol 2012; 119: 396-407. Fujita Y, Tanimura M, Tanaka K: The distribution of the ABO blood groups in Japan. Jinrui Idengaku Zasshi 1978; 23: 63-109. Ben-Haroush A, Yogev Y, Hod M: Epidemiology of gestational diabetes mellitus and its association with Type 2 diabetes. Diabet Med 2004; 21: 103-113. Seyfizadeh N, Yousefi B, Borzoueisileh S, Majidinia M, Shanehbandi D, Jahani MA: Is there association between ABO blood group and the risk factors of unfavorable outcomes of pregnancy? J Matern Fetal Neonatal Med 2015; 28: 578-582. Okuda H, Kawano M, Iwamoto S, Tanaka M, Seno T, Okubo Y, Kajii E: The RHD gene is highly detectable in RhD-negative Japanese donors. J Clin Invest 1997; 100: 373-379. Liu YH, Koda Y, Soejima M, Pang H, Wang BJ, Kim DS, Oh HB, Kimura H: The fusion gene at the ABO-secretor locus (FUT2): absence in Chinese populations. J Hum Genet 1999; 44: 181-184. Melzer D, Perry JR, Hernandez D, Corsi AM, Stevens K, Rafferty I, Lauretani F, Murray A, Gibbs JR, Paolisso G, Rafiq S, Simon-Sanchez J, Lango H, Scholz S, Weedon MN, Arepalli S, Rice N, Washecka N, Hurst A, Britton A, Henley W, 14 Page 14 of 18

ip t

cr

us an M d Ac ce pt e

24.

van de Leemput J, Li R, Newman AB, Tranah G, Harris T, Panicker V, Dayan C, Bennett A, McCarthy MI, Ruokonen A, Jarvelin MR, Guralnik J, Bandinelli S, Frayling TM, Singleton A, Ferrucci L: A genome-wide association study identifies protein quantitative trait loci (pQTLs). PLoS Genet 2008; 4: e1000072. Paterson AD, Lopes-Virella MF, Waggott D, Boright AP, Hosseini SM, Carter RE, Shen E, Mirea L, Bharaj B, Sun L, Bull SB: Genome-wide association identifies the ABO blood group as a major locus associated with serum levels of soluble E-selectin. Arterioscler Thromb Vasc Biol 2009; 29: 1958-1967.

15 Page 15 of 18

non-GDM

GDM

5275

149

30.7±4.9 2184 (41.4) 1885 (35.7) 1206 (22.9)

33.6±5.4 31 (20.8) 51 (34.2) 67 (45.0)

159.1±4.0

157.9±6.4

Pre-pregnancy BMI (kg/m )

20.7±5.5

23.8±5.5

ip t

Table 1. Clinical and biochemical characteristics of subjects with non-GDM and GDM.

< 0.001

a

Random glucose level (mg/dL)

80.9±7.0

85.4±14.3

< 0.001

a

Parity ≥1 (%)

2748 (52.1)

99 (66.4)

< 0.001

b

Singleton pregnancy (%)

5117 (97.0)

< 30 (%) 31-34 (%) ≥35 (%) Body height (cm) 2

< 0.001

< 0.001

b

149 (100.0)

0.011

19 (12.8)

< 0.001

b

28 (18.8)

< 0.001

b

Familial history of diabetes (%)

301 (5.7)

Smoker (%)

232 (4.4)

8 (5.4)

0.330

Alcohol drinker (%)

380 (7.2)

7 (4.7)

0.157

1999 (37.9) 1250 (23.7) 1507 (28.6) 519 (9.8)

45 (30.2) 35 (23.5) 40 (26.8) 19 (19.5)

0.001

5250 (99.5) 25 (0.5)

149 (100.0) 0 (0.0)

0.400

d

M

an

132 (2.5)

Ac ce pt e

b

a

Hypertenison (%)

ABO blood type (%) A B O AB Rh blood type (%) Positive Negative

a

0.755

cr

Age (years)

us

Number of subjects

P

b b

b

b

GDM; gestational diabetes mellitus, BMI; body mass index Data are shown in mean ± standard deviation or number (%). a

Derived from Student's t-test.

b

Derived from χ test or Fisher's exact test.

2

16 Page 16 of 18

cr us an

Table 2. Clinical and biochemical characteristics of subjects by ABO blood type. Blood type

B

O

AB

2044 (37.7)

1285 (23.7)

1547 (28.5)

548 (10.1)

30.8±4.9 833 (40.8) 724 (35.4) 487 (23.8)

30.9±5.0 492 (38.3) 484 (37.7) 309 (24.0)

30.5±5.0 676 (43.7) 531 (34.3) 340 (22.0)

30.8±5.0 214 (39.1) 197 (35.9) 137 (25.0)

0.180

a

0.067

b

160.0±4.1

157.9±7.4

157.9±7.3

161.4±6.8

0.260

a

20.6±6.1

21.1±7.4

20.8±3.0

20.5±3.5

0.153

a

81.3±7.3

81.0±7.1

80.9±7.1

81.0±8.6

0.566

Parity ≥1 (%)

1049 (51.3)

698 (54.3)

803 (51.9)

298 (54.4)

0.290

b

Singleton pregnancy (%)

1970 (96.4)

1252 (97.0)

1515 (97.9)

534 (97.4)

0.062

b

Hypertenison (%)

56 (2.7)

42 (3.3)

41 (2.7)

13 (2.4)

0.669

b

Familial history of diabetes (%)

121 (5.9)

67 (5.2)

103 (6.7)

36 (6.6)

0.405

Smoker (%)

92 (4.5)

55 (4.3)

63 (4.1)

28 (5.1)

0.766

b

Alcohol drinker (%)

142 (6.9)

81 (6.3)

114 (7.4)

49 (8.9)

0.232

b

M

A

Number of subjects (%)

Body height (cm) 2

BMI (kg/m )

pt

< 30 (%) 31-34 (%) ≥35 (%)

ed

Age (years)

Ac

ce

Random glucose level (mg/dL)

Data are shown in mean ± standard deviation or number (%). a

Derived from Student's t-test.

b

Derived from χ test or Fisher's exact test.

2

17 Page 17 of 18

P

a

b

cr us an

Table 3. Association between ABO blood type and GDM.

Model M1

Model M0 OR (95% CI) 0.42 (0.26-0.66)

B

0.52 (0.32-0.83)

O

0.49 (0.31-0.78)

P

OR (95% CI)

P

OR (95% CI)

P

<0.001

0.35 (0.20-0.62)

<0.001

0.35 (0.20-0.63)

<0.001

0.34 (0.19-0.63)

0.001

0.006

0.23 (0.32-0.78)

0.006

0.39 (0.21-0.74)

0.004

0.35 (0.18-0.68)

0.006

0.003

0.41 (0.23-0.75)

0.004

0.45 (0.25-0.82)

0.010

0.39 (0.21-0.74)

0.004

ed

A

Model M3

Model M2

OR (95% CI)

P

M

Blood type

1.00 (reference)

1.00 (reference)

1.00 (reference)

1.00 (reference)

non-AB

1.00 (reference)

1.00 (reference)

1.00 (reference)

1.00 (reference)

2.23 (1.46-3.36) < 0.001

2.56 (1.57-4.18) < 0.001

2.61 (1.58-4.30) < 0.001

2.73 (1.64-4.57) < 0.001

AB

ce

M0, crude model

pt

AB

M1, adjusted for maternal age and BMI. M2, adjusted for maternal age, BMI, parity, and hypertension.

Ac

M3, adjusted for maternal age, BMI, parity, hypertension, non-singleton pregnancy, and familial history of diabetes.

18 Page 18 of 18