Age at menarche and age at natural menopause as predictors of glycemic control in type 2 diabetic patients

Accepted Manuscript Age at menarche and age at natural menopause as predictors of glycemic control in type 2 diabetic patients

Chong Shen, Jinbo Wen, Xiaoqun Pan, Jian Su, Wencong Du, Enchun Pan, Qin Zhang, Ning Zhang, Hongyan Sheng, Chunlan Liu, Hui Wang, Ming Wu, Yu Qin PII: DOI: Reference:

S1056-8727(18)30121-1 doi:10.1016/j.jdiacomp.2018.04.012 JDC 7204

To appear in: Received date: Revised date: Accepted date:

31 January 2018 20 March 2018 24 April 2018

Please cite this article as: Chong Shen, Jinbo Wen, Xiaoqun Pan, Jian Su, Wencong Du, Enchun Pan, Qin Zhang, Ning Zhang, Hongyan Sheng, Chunlan Liu, Hui Wang, Ming Wu, Yu Qin , Age at menarche and age at natural menopause as predictors of glycemic control in type 2 diabetic patients. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jdc(2018), doi:10.1016/ j.jdiacomp.2018.04.012

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.

ACCEPTED MANUSCRIPT Age at Menarche and Age at Natural Menopause as Predictors of Glycemic Control in Type 2 Diabetic Patients Chong Shen1,+, Jinbo Wen3,+, Xiaoqun Pan2, Jian Su2, Wencong Du2, Enchun Pan3, Qin Zhang3, Ning Zhang4, Hongyan Sheng4, Chunlan Liu1, Hui Wang1, Ming Wu2,*, Yu Qin2,*

PT

1. Department of Epidemiology, School of Public Health, Nanjing Medical University, Nanjing 211166, China

RI

2. Department of Non-communicable Chronic Disease Control, Jiangsu Provincial Center for Disease Control and Prevention, Nanjing, 210009, China

SC

3. Department of Chronic Disease Prevention and Control, Huai’an City Center for Disease Control and Prevention, Huai’an, 223001, China

NU

4. Changshu County Center for Disease Control and Prevention, Suzhou, 215500,

MA

China

Wu,

Tel:

+86

25

83759310,

PT E

Ming

D

To whom correspondence should be addressed: Fax:

+86

25

83759310,

E-mail:

[email protected]; Yu Qin, Tel: +86 25 83759443, Fax: +86 25 83759443, E-mail: [email protected].

AC

CE

Chong Shen and Jinbo Wen contributed equally to this work.

1

ACCEPTED MANUSCRIPT Abstract Objective To investigate the influence of age at menarche (AM) and age at natural menopause (ANM) on glycemic control in patients with type 2 diabetes mellitus (T2DM). Methods A cross-sectional study was undertaken in Jiangsu, China. Logistic regression was used to evaluate the association between AM/ANM and glycemic

PT

control. Results 1,195 (14.3%) premenopausal and 7,149 (85.7%) postmenopausal women were included in this study. With the increase of AM per 1 year, patients had a

RI

low risk of uncontrolled FPG (7 mmol/L) and uncontrolled HbA1c (7%), as well as poor glycemic control (FPG 7 mmol/L and HbA1c 7%) after adjusting for age and

SC

BMI (model I, P <0.05) with odds ratio (OR) 0.965, 0.978 and 0.962 respectively. Whereas after adjusting for age, BMI, education, duration of diabetes, smoking,

NU

drinking and antidiabetic treatment (model II) as well as further plus diabetic familial history and physical activity (model III), the association between AM and glycemic

MA

control was not significant (P >0.05). Compared with premenopausal women, postmenopausal women had a low risk of uncontrolled FPG and uncontrolled HbA1c

D

after adjusting for confounders in model II (P <0.05). Furthermore, both patients with

PT E

early ANM (<45 years) and late ANM (>55 years) had a high risk of uncontrolled HbA1c as well as poor glycemic control even adjusted for full confounders in model III (P <0.05 for all). Conclusion Early AM, early and late ANM were significantly

CE

associated with worse glycemic control. Ascertaining the AM and ANM in women

AC

with T2DM may help to identify the risk predisposed to worse glycemic control.

Key words: Type 2 diabetes mellitus; Menarche; Menopause; Fasting plasma glucose; Hemoglobin A1c

2

ACCEPTED MANUSCRIPT Introduction Approximately 11.0% of women over 18 years old were estimated to live with diabetes, which meant a total of 53.4 million women had diabetes in China in 20101. Diabetes, as one of the most concerned non-communicable diseases, is also a growing cause of disability and premature death in China 2. Moreover, diabetes and its

PT

complications such as coronary heart disease and stroke pose threats to the patient’s health and increase the health care burden to the patients, their family, and the society

RI

as a whole3, 4. It is estimated that, in general, patients with diabetes aged 55-74 years have a reduction in life expectancy of about 4-8 years 5.A pooled analysis of a

SC

population with type 2 diabetes mellitus (T2DM) in Germany, indicates that T2DM significantly deteriorates the quality of life of women, especially to mental health 6.

NU

Diabetes was found to be significantly positively associated with urgency incontinence in a United States study of 121,701 women 7, in a Norwegian study of

MA

21,057 women 8, and in a US study of 2,763 postmenopausal women 9. Maintaining adequate blood glucose is important for T2DM patients that each 1% reduction in

D

HbA1c was associated with a 37% decrease in risk for microvascular complications

PT E

and a 21% decrease in the risk of any end point or death related to diabetes10.

CE

Menarche and menopause are landmarks that signal the beginning and end of normal reproductive life for women. Age at menarche (AM) and age at natural menopause

AC

(ANM) have been reported associated with many health related diseases for women in later life

11

. It was reported that AM is associated with risks for obesity, type 2

diabetes, cardiovascular disease, breast cancer and all-cause mortality12, while there was no consensus about the association between AM and diabetes mellitus. Some research found no association between AM and T2DM13, while other study indicated that early menarche is associated with increased risk of T2DM in adulthood14.

The menopausal transition is a period of rapid change in physiologic body composition and body fat distribution, and lipid and metabolic profiles 15. Thus, it has 3

ACCEPTED MANUSCRIPT been hypothesized that these underlying changes represent a mechanistic link between diabetes and menopause16. Previously cross-sectional and longitudinal studies have evaluated the relationship between menopausal status and diabetes risk while the results were controversial, some results indicated that postmenopausal women had a significant association with T2DM17, while in other research no association was found

PT

between natural menopause and diabetes18. Moreover, whether age at menopause is associated with diabetic risk has also been investigated with conflicting results,

RI

prospective cohort studies found that early age at menopause had a high risk of

SC

incident diabetes19, 20, while other researchers identified no association 21, 22.

As before, epidemiological evidence on the association between reproductive factors

NU

and glycemic control in T2DM patients is apparently rare. Understanding the relationships between ages at menarche and menopause and glycemic control may

MA

lead to more effective prevention and management of the T2DM and its complications. Herein, we mainly aimed at investigating whether AM, menopause status, ANM are

D

associated with glycemic control in women with T2DM by a cross-sectional study of

PT E

Chinese population.

Subject

CE

Subjects were from the project of “Comprehensive Research on the Prevention and Control of the Diabetes (CRPCD)”. The method and study design have been

AC

previously described

23

. According to the criteria established by the American

Diabetes Association (ADA) in 2010 24, T2DM patients were defined as having FPG levels 7.0mmol/L or having a self-reported T2DM history without type 1 diabetes mellitus. A total of 20,340 subjects agreed to complete a standardized epidemiology questionnaire and physical examination. Ultimately 19,992 T2DM patients, which consist of 7843 (39.2%) men and 12149 (60.8%) women, had both the FPG and HbA1c detections. Men were excluded from this study, as were women who met the following criteria: (1) postmenopausal women with hormone replacement therapy (HRT) or the information about HRT was missing (n=800); (2) with non-natural 4

ACCEPTED MANUSCRIPT menopause (surgical, or other reason) or not report a cause for menopause (n=875); (3) onset age of T2DM was earlier than ANM or information of onset age was missing (n=1790); (4) information of menopausal status was missing or age at menopause in postmenopausal women was missing (n=179); (5) information of AM was missing

PT

(n=161). Finally, 8344 women were included in this study.

The research protocol of this study was reviewed and approved by the Ethics Board of

RI

Jiangsu Provincial Center for Disease Control and Prevention (No. 2013026). All patients were well informed about the study and provided written consent before

SC

participating and all methods were performed in accordance with the relevant

NU

guidelines and regulations.

Questionnaire survey, anthropometric measurements and biochemical indices

MA

detection

The investigators were trained and qualified in a standardized manner. Demographic

D

characteristics, medical history, and lifestyle factors of the patients were collected by

PT E

a standardized questionnaire. Question about age at menarche was: “When is the age of menarche?” Question about menopause status was: “Are you menopause now?” Question about age at menopause was: “The age when your menopause?”

CE

Question about reason of menopause was: “The reason of menopause: natural; operation; others?” Question about hormone replacement treatment (HRT) was:

AC

“Whether do you have HRT treatment?” Anthropometric measurements, including height (meter) and body weight (kilogram), were obtained by trained technicians. Height and body weight was measured twice for individuals ensuring that heavy clothes and shoes were removed before measurements, and the figures were rounded as precisely as 0.1 cm and 0.1 kg, respectively. Body mass index (BMI) was calculated as weight (kilograms)/height (meters)2. Blood samples were collected in the morning after an overnight fast. Serum fasting glucose was assessed using the enzymatic method, and the whole blood HbA1c level was assessed using commercially available reagents from KingMed Diagnostics (Jiangsu Cultural 5

ACCEPTED MANUSCRIPT Industrial Park, Nanjing, China).

Definition Assessment indices for glycemic control of T2DM were categorized as FPG: <7 mmol/L (controlled) and ≥7 mmol/L (uncontrolled); HbA1c: <7% (controlled) and ≥7% (uncontrolled) 25. In addition, an evaluation of HbA1c and FPG combined was used to

PT

assess glycemic control status. The status of glycemic control were defined as “good”,

RI

“common” and “poor” when the criteria of HbA1c <7% and FPG <7 mmol/L were

SC

both met, when either criterion was met, and when neither was met, respectively.

Characteristics were defined as follows. Smoking habits: have ever smoked 100

NU

cigarettes in the past; drinking: at least once alcoholic drink per month. FH of diabetes were recorded when the patient’s father, mother, sibling, grandparents/maternal

MA

grandparents, uncles, and/or aunts were reported to have a diabetic history. Metabolic equivalents (METs) were used to assess the physical activity level and the detail 26

. Natural menopause is diagnosed after 12

D

method has been previously illustrated

PT E

months of amenorrhea not associated with a pathological cause27. AM was categorized as: <15, 16-18 and >18 years and ANM was classified as: <45,

CE

45-55and >55 years.

Statistical analysis

AC

The mean and standard deviation (SD) was presented for continuous variables with a normal distribution. Whereas those with a skewed distribution were presented with the median and the interquartile range. A chi-square (2) test was used for the comparisons of categorical variables. Student’s t-test and analysis of variance (ANOVA) were used for to compare the normally-distributed continuous variables and the Mann-Whitney U test was used for to compare abnormally-distributed variables. Logistic regression analyses were performed, and the relative risk was estimated as the odds ratio (OR) and 95% confidence interval (CI). A two-tailed P value of <0.05 was considered as statistically significant. All statistical analyses were 6

ACCEPTED MANUSCRIPT performed using IBM SPSS Statistics 15.0 (SPSS, Inc, Chicago, USA).

Three models of logistic regression analysis were performed for the simultaneous assessment of various variables influencing glycemic control (dependent variable). In the regression model I, only age and BMI were adjusted. Age, BMI, education,

PT

duration of diabetes, smoking, drinking and antidiabetic treatment were further adjusted in model II. The regression model III included familial history of diabetes

RI

and energy metabolism as well as those in model II. In each logistic regression analysis, variables of ‘AM’, ‘category of AM’, ‘menopausal status’, ‘ANM’ and

SC

‘category of ANM’ were entered in models separately to estimate the association with glycemic control.

NU

Role of the founding source

The funders of this study had no role in study design, data collection and analysis,

MA

decision to publish, or preparation of the manuscript. The corresponding author had full access to all the data in the study and had final responsibility for the decision to

PT E

D

submit for publication.

Results

Demographic and basic characteristics of study population

CE

In the total 8,344 subjects, 1,195 (14.3%) were premenopausal women and 7,149 (85.7%) were postmenopausal women. The mean age of the whole population,

AC

premenopausal women and postmenopausal women were 64.59, 48.23 and 67.32 years respectively. Median duration of diabetes in postmenopausal women was significantly higher than that of premenopausal women (4.17 years and 3.42 years respectively) with P value <0.001. The mean levels of FPG and HbA1c in premenopausal women were significantly higher than those of postmenopausal women with 9.28 mmol/L, 8.40 mmol/L respectively (P <0.001) and 7.80%, 7.48% (P <0.001) respectively. The proportion of subjects with “good” glycemic control according to FPG & HbA1c index in postmenopausal women (30.4%) was significantly higher than that of premenopausal women (27.8%), and P value <0.001. 7

ACCEPTED MANUSCRIPT The detail data were listed in supplementary table 1.

General characteristics of participants with category of AM and ANM The numbers of subject with AM of “<15”, “16-18” and “>18” years were 1304 (15.6%), 5576 (66.8%) and 1464 (17.5%), respectively. Age, BMI and duration of

PT

diabetes were significantly different between these three groups categorized by AM with P value all <0.05. Furthermore, with the increase of AM grade, average age

RI

increased gradually (60.86 years, 65.00 years, and 66.35 years respectively) and BMI decreased gradually (25.64 kg/m2, 25.37 kg/m2, and 25.18 kg/m2 respectively). Thus,

SC

BMI and age would be adjusted to evaluate the influence of AM on glycemic control in the multiple regression analysis. No significant difference of physical activity was

NU

found between these three groups (P >0.05). Subjects with AM >18 years had higher proportions of smoking, drinking and without formal education, and subjects with AM

MA

<15 years had higher proportion of familial history of diabetes (P values all <0.05), compared with subjects with AM 15-18 years. Furthermore, with the increase of AM

D

grade, the proportions of smoking and without formal education increased gradually,

PT E

with 5.7%, 8.4%, and 13.1% respectively (P <0.001) and 41.8%, 54.4%, and 72.1% (P<0.001) respectively. Conversely, the proportion of familial history of diabetes decreased gradually with 25.2%, 19.6% and 14.7% (P <0.001). The detail information

CE

was presented in table 1.

AC

The number of subjects with ANM of “<45”, “45-55” and “>55” years was 601 (8.4%), 6178 (86.4%) and 370 (5.2%), respectively. There was no significant difference between these three groups regarding to smoking, drinking, education, physical activity and familial history of diabetes (P values all >0.05). BMI increased gradually with the increase of ANM grade (24.95 kg/m2, 25.34 kg/m2 and 25.73 kg/m2, P=0.003), while duration of diabetes and proportion of antidiabetic treatment decreased gradually with the increase of ANM grade with P values all <0.05 (4.50 years, 4.17 years and 3.67 years, and 80.6%, 75.9 % and 75.1% respectively). In addition, AM was higher among subjects of ANM >55 years than subjects of ANM 8

ACCEPTED MANUSCRIPT <45 years and AMN 45-55 years (17.02 years, 16.94 years, and 16.80 years, P=0.045). (Table 2)

Table 1: General characteristics of participants with category of AM Category of AM

CE

Physical activity (MET- hour/day)

Familial history of diabetes (%) Yes No

AC

>18 years

1304 (15.6) 60.86 ± 12.13 25.64 ± 3.78 4.00 (1.75, 7.42)

5576 (66.8) 65.00 ± 9.94 25.37 ± 3.52 4.17 (1.75, 7.58)

1464 (17.5) 66.35 ± 7.66 25.18 ± 3.54 3.58 (1.50, 6.73)

74 (5.7) 1230 (94.3)

471 (8.4) 5105 (91.6)

192 (13.1) 1272 (86.9)

<0.001

29 (2.2) 14 (1.1) 1257 (96.7)

116 (2.1) 53 (1.0) 5384 (97.0)

44 (3.0) 27 (1.8) 1391 (95.1)

0.012

544 (41.8) 435 (33.5) 231 (17.8) 90 (6.9)

3018 (54.4) 1737 (31.3) 589 (10.6) 208 (3.7)

1051 (72.1) 329 (22.6) 59 (4.0) 19 (1.3)

<0.001

983 (75.8) 313 (24.2)

4270 (77.2) 1264 (22.8)

1053 (72.5) 399 (27.5)

0.001

31.50 (27.78, 37.30)

31.37 (27.77, 36.42)

31.30 (27.85, 36.85)

0.423

314 (25.2) 931 (74.8)

1037 (19.6) 4248 (80.4)

206 (14.7) 1198 (85.3)

<0.001

MA

NU

SC

RI

PT

15-18 years

PT E

No. participants (%) Age (year) BMI (kg/m2) Duration of diabetes (year) Smoking (%) Yes No Drinking (%) Yes Former No Education (%) Without formal education Primary and below Middle school High school and above Antidiabetic treatment (%) Yes No

P

<15 years

D

Variable

<0.001 0.003 <0.001

Continuous variables were presented as means (SD) and categorized variables as numbers (percentages). There were 11 missing values for BMI, 29 for drinking, 34 for education, 62 for antidiabetic treatment, 568 for physical activity, and 410 for familial history of diabetes.

9

ACCEPTED MANUSCRIPT

Table 2: General characteristics of participants with different category of ANM Category of ANM

Physical activity (MET- hour/day)

>55 years

601 (8.4) 67.70 ± 8.62 24.95 ± 3.52 4.50 (2.00, 8.42)

6178 (86.4) 67.22 ± 7.44 25.34 ± 3.55 4.17 (1.75, 7.67)

370 (5.2) 68.36 ± 6.85 25.73 ± 3.63 3.67 (1.50, 6.33)

69 (11.5) 532 (88.5)

572 (9.3) 5606 (90.7)

12 (2.0) 5 (0.8) 582 (97.2)

149 (2.4) 73 (1.2) 5935 (96.4)

CE

Familial history of diabetes (%) Yes No AM (year)

PT

45-55 years

0.008 0.003 <0.001

0.195

RI

37 (10.0) 333 (90.0)

P

0.795

3747 (60.9) 1740 (28.3) 479 (7.8) 188 (3.1)

228 (62.1) 99 (27.0) 26 (7.1) 14 (3.8)

0.060

478 (80.6) 115 (19.4)

4657 (75.9) 1479 (24.1)

275 (75.1) 91 (24.9)

0.032

31.61 (27.54, 37.21)

31.20 (27.67, 35.85)

31.70 (28.01, 37.20)

0.085

104 (18.3) 465 (81.7) 16.94 ± 2.24

1044 (17.8) 4822 (82.2) 16.80 ± 2.04

55 (15.4) 302 (84.6) 17.02 ± 2.13

NU

MA

399 (66.6) 155 (25.9) 28 (4.7) 17 (2.8)

SC

9 (2.4) 6 (1.6) 354 (95.9)

PT E

No. participants (%) Age (year) BMI (kg/m2) Duration of diabetes (year) Smoking Yes No Drinking (%) Yes Former No Education (%) Without formal education Primary and below Middle school High school and above Antidiabetic treatment (%) Yes No

<45 years

D

Variable

0.483 0.045

AC

Continuous variables were presented as means (SD) and categorized variables as numbers (percentages). There were 11 missing values for BMI, 24 for drinking, 29 for education, 54 for antidiabetic treatment, 465 for physical activity, and 357 for familial history of diabetes.

Glycemic control among different categories of AM and ANM Among AM of “<15”, “15-18” and “>18” years groups, the levels of FPG and HbA1c displayed no significant difference (P >0.05, but the proportions of glycemic control according to FPG <7 mmol/L (34.7 %, 35.8 %, and 40.1%) and HbA1c <7% (43.6%, 46.1% and 48.0%) both increased gradually, and P values for trend were 0.002 and 10

ACCEPTED MANUSCRIPT 0.021 respectively. The proportion of “good” glycemic status based on FPG & HbA1c also increased across the three groups (27.8%, 29.8 % and 32.9%) and P value for trend was 0.003 (Table 3).

Subjects with ANM of “45-55” years had a significantly lower FPG level and HbA1c

PT

level than subjects with ANM of “<45” and “>55” years, and the levels of FPG and HbA1c were 8.35 mmol/L, 8.78 mmol/L and 8.58 mmol/L, and 7.45 %, 7.69 % and

RI

7.52% respectively (both P values were 0.004). Accordingly, the proportions of glycemic under control with FPG <7 mmol/L and HbA1c <7% were significantly

SC

higher in ANM of “45-55” years group (37.9 % and 47.5%) than that in ANM of “<45” and “>55” years groups (31.1 %, 34.1% and 41.6 %, 44.6%), and P value were 0.002

NU

and 0.014 respectively. Similarly, the proportion of “good” glycemic status for FPG & HbA1c index in the group of ANM “45-55” years (31.0%) was significantly higher

MA

than that in groups of ANM “<45” (26.0%) and “>55” (27.0%) years, and P value was

AC

CE

PT E

D

0.004 (Table 3).

11

ACCEPTED MANUSCRIPT

Table 3: Comparison of glycemic control between different categories of AM and ANM Variable

Category of AM

45-55 years

>55 years

0.241 (0.106) 0.370 (0.302)

8.78 ± 3.64 7.69 ± 1.76

8.35 ± 3.11 7.45 ± 1.67

8.58 ± 3.20 7.52 ± 1.71

0.004 (0.356) 0.004 (0.118)

587 (40.1) 877 (59.9)

0.004 (0.002)

187 (31.1) 414 (68.9)

2341 (37.9) 3837 (62.1)

126 (34.1) 244 (65.9)

0.002 (0.099)

2573 (46.1) 3003 (53.9)

702 (48.0) 762 (52.0)

0.067 (0.021)

250 (41.6) 351 (58.4)

2937 (47.5) 3241 (52.5)

165 (44.6) 205 (55.4)

0.014 (0.132)

647 (49.6)

2665 (47.8)

656 (44.8)

320 (53.2)

2816 (45.6)

179 (48.4)

Common

294 (22.5)

1251 (22.4)

327 (22.3)

125 (20.8)

1446 (23.4)

91 (24.6)

Good

363 (27.8)

1660 (29.8)

481 (32.9)

156 (26.0)

1916 (31.0)

100 (27.0)

7 mmol/L HbA1c category (%) <7% 7% FPG & HbA1c (%) Poor

15-18 years

>18 years

8.61 ± 3.29 7.59 ± 1.77

8.54 ± 3.31 7.51 ± 1.71

8.40 ± 3.30 7.52 ± 1.81

452 (34.7) 852 (65.3)

1998 (35.8) 3578 (64.2)

568 (43.6) 736 (56.4)

P (Ptrend)

<45 years

FPG (mmol/L) HbA1c (%) FPG category (%) <7 mmol/L

<15 years

Category of ANM

P (Ptrend)

D E

A M

0.046 (0.003)

T P E

C C

A

12

I R

SC

U N

T P

0.004 (0.072)

ACCEPTED MANUSCRIPT Evaluation for the association of AM and ANM with FPG control Multiple logistic regression analysis showed that after adjusting for age and BMI in model I, AM was significantly associated with a low risk of FPG 7 mmol/L (uncontrolled) with adjusted OR (95%CI) of 0.965 (0.944-0.986) and P value of 0.001, nevertheless, the association was attenuated after adjustment for other variables in

PT

model II and model III with adjusted ORs (95%CIs) of 0.980 (0.957-1.004) and 0.980 (0.955-1.006), respectively. With the increase of AM grade (<15 years,

15-18 years

RI

and >18 years), subjects had a decreased risk of uncontrolled FPG with adjusted OR (95%CI) of 0.886 (0.819-0.958) and Ptrend value of 0.002 after adjustment for age and

SC

BMI. While the association was not statistically significant in model II and model III

respectively. (Supplementary table 2).

NU

with adjusted ORs (95%CIs) of 0.928 (0.850-1.013) and 0.917 (0.835-1.007),

MA

Compared with premenopausal subjects, postmenopausal subjects had a low risk of uncontrolled FPG with ORs (95%CIs) of 0.670 (0.563-0.796) and 0.841 (0.671-0.987)

D

respectively after adjustment for confounders in model I and model II. While further

PT E

adjusting for other confounders in model III, the adjusted OR (95%CI) was 0.818 (0.664-1.007) with a marginal P value of 0.058. No significant association was observed between ANM grade and uncontrolled FPG status in all of the three models.

CE

Compared with subjects with ANM of “45-55”, subjects with ANM of “<45” years had a high risk of uncontrolled FPG and adjusted ORs (95%CI) were 1.344

AC

(1.122-1.609) and 1.268 (1.039-1.548) respectively after adjustment for confounders in model I and model II. While in model III, the association presented adjusted OR (95%CI) 1.229 (0.995-1.517) with a marginal P value of 0.055. Subjects with ANM of “>55” years had a significantly high risk of uncontrolled FPG when compared with ANM of “45-55” years even after adjustment for confounders in model II and model III, and adjusted ORs (95%CIs) were 1.363 (1.069-1.739) and 1.407 (1.089-1.817) respectively (Supplementary table 2). Evaluation for the association of AM and ANM with HbA1c control After adjustment for age and BMI, AM was significantly associated with a low risk of 13

ACCEPTED MANUSCRIPT HbA1c 7% (uncontrolled) with adjusted OR (95%CI) of 0.978 (0.958-0.999) and adjusted P value of 0.040 while the association was not significant in model II and model III with adjusted ORs (95%CIs) of 0.997 (0.974-1.021) and 1.000 (0.975-1.025), respectively. The increase of AM grade was significantly associated with lower risk of uncontrolled HbA1c with adjusted OR (95%CI) of 0.923

PT

(0.856-0.996) and Ptrend value of 0.039 after adjusting age and BMI. While the association was not significant in model II and model III with adjusted ORs (95%CIs)

RI

of 0.978 (0.898-1.064) and 0.974 (0.890-1.066), respectively. No significant

all of the three models with P values all >0.1.

SC

association was found between groups categorized by AM and uncontrolled HbA1c in

NU

Compared with premenopausal subjects , postmenopausal subjects had a significantly low risk of uncontrolled HbA1c even after adjustment for confounders in Model III

MA

and adjusted OR was 0.811 (P=0.039). Both subjects with ANM of “<45” and “>55” years had a high risk of uncontrolled HbA1c when compared with subjects with ANM

D

of “45-55” years, and adjusted ORs (95%CI) were 1.220 (1.000-1.488) and 1.306

PT E

(1.025-1.664) respectively after adjustment for confounders in model III (Supplementary table 3).

CE

Evaluation for the association of AM and ANM with combined glycemic control of FPG & HbA1c index

AC

Results of multiple logistic regression analysis of the association between AM and ANM with glycemic control status (poor glycemic control and common glycemic control compared with good glycemic control) based on FPG & HbA1c were presented in Table 4 and Table 5. With the increase of AM per 1 year, subjects had a low risk of poor glycemic control and adjusted OR (95%CI) was 0.962 (0.939-0.986) after adjustment for age and BMI while the significance did not persist after adjustment for other variables in model II and model III with adjusted ORs (95%CIs) of 0.984 (0.957-1.012) and 0.984 (0.955-1.014), respectively. The increase of AM grade was significantly associated with lower risk of poor glycemic control with 14

ACCEPTED MANUSCRIPT adjusted OR (95%CI) of 0.874 (0.801-0.955) after adjustment for age and BMI while the association trend was not significant in model II and model III with adjusted ORs of (95%CIs) 0.933 (0.843-1.033) and 0.919 (0.825-1.024), respectively. Significantly, subjects with AM of “>18” years had a low risk of poor glycemic control with adjusted OR (95% CI) of 0.854 (0.747-0.975) and adjusted P value of 0.020 in model

PT

I, compared with subjects with AM of “15-18” years. Postmenopausal women had a significantly low risk of poor glycemic control when

RI

compared with premenopausal women with adjusted ORs (95% CIs) of 0.602 (0.497-0.730) and 0.789 (0.633-0.983) respectively in model I and model II, while in

SC

model III, the association presented OR (95%CI) of 0.790 (0.623-1.003) with a marginal P value of 0.052. No significant association trend was observed between

NU

ANM grade and poor glycemic control in all three models (P value all >0.1). Subjects with ANM of “<45” and “>55” years had a high risk of poor glycemic control when

MA

compared with subjects with ANM of “45-55” years and adjusted ORs (95%CIs)

D

1.286 (1.009-1.639) and 1.565 (1.157-2.116) respectively in model III (Table 4).

PT E

With the increase of AM per 1 year, subjects had a low risk of common glycemic control and OR (95%CI) was 0.967 (0.939-0.996) after adjustment for age and BMI, while the significance did not persist after adjustment for other variables in model II

CE

and model III with adjusted ORs (95%CI) of 0.987 (0.956-1.019) and 0.979 (0.947-1.012), respectively. The increase of AM grade was significantly associated

AC

with lower risk of common glycemic control and OR (95%CI) was 0.889 (0.800-0.988) after adjustment for age and BMI. In terms of the association between groups categorized by AM and common glycemic control, no statistical significance was found in all of the three models with P values significantly greater than 0.05. The risk of common glycemic control did not differ between premenopausal women and postmenopausal women in different models, and the risk did not differ with the increase of ANM as well as ANM grade, and P values all >0.1. Subjects with ANM of “>55” years had a high risk of common glycemic control, relative to subjects with ANM of “45-55” years in model III, and OR (95%CI) was 1.442 (1.039-2.001) with P 15

ACCEPTED MANUSCRIPT value 0.029(Table 5).

Table 4: The association between menarche and menopause and glycemic control regarding to FPG & HbA1c index (poor vs good) Model I

Model III

P

OR (95%CI)

P

OR (95%CI)

P

0.962 (0.939-0.986) 0.874 (0.801-0.955) 1 1.113 (0.964-1.286) 0.854 (0.747-0.975)

0.002 0.003 1 0.144 0.020

0.984 (0.957-1.012) 0.933 (0.843-1.033) 1 1.144 (0.969-1.349) 0.987 (0.846-1.151)

0.264 0.184 1 0.112 0.869

0.984 (0.955-1.014) 0.919 (0.825-1.024) 1 1.115 (0.936-1.328) 0.937 (0.795-1.104)

0.294 0.127 1 0.224 0.436

1 0.602 (0.497-0.730) 0.990 (0.976-1.003) 0.867 (0.748-1.006) 1 1.404 (1.149-1.716) 1.175 (0.931-1.512)

1 <0.001 0.134 0.060 1 0.001 0.211

1 0.789 (0.633-0.983) 1.004 (0.989-1.020) 0.998 (0.843-1.180) 1 1.295 (1.030-1.629) 1.467 (1.103-1.951)

1 0.790 (0.623-1.003) 1.003 (0.986-1.019) 1.029 (0.862-1.427) 1 1.286 (1.009-1.639) 1.565 (1.157-2.116)

1 0.052 0.748 0.752 1 0.042 0.004

SC

RI

PT

OR (95%CI)

MA

AM (year) Category of AM 15-18 years <15 years >18 years Menopause status Pre-menopause Post-menopause ANM (year) Category of ANM 45-55 years <45 years >55 years

Model II

NU

Variable

1 0.035 0.612 0.979 1 0.027 0.008

AC

CE

PT E

D

Model I adjusted for age and BMI. Model II adjusted for age, BMI, education, duration of diabetes, drinking, smoking, and antidiabetic treatment. Model III adjusted for age, BMI, education, duration of diabetes, drinking, smoking, antidiabetic treatment, familial history of diabetes, and physical activity.

16

ACCEPTED MANUSCRIPT

Table 5: The association between menarche and menopause and glycemic control regarding to FPG & HbA1c index (common vs good) Model I

Model III

P

OR (95%CI)

P

OR (95%CI)

P

0.967 (0.939-0.996) 0.889 (0.800-0.988) 1 1.128 (0.949-1.341) 0.893 (0.762-1.047)

0.024 0.029 1 0.171 0.162

0.987 (0.956-1.019) 0.944 (0.842-1.058) 1 1.164 (0.968-1.401) 1.025 (0.863-1.217)

0.423 0.319 1 0.107 0.780

0.979 (0.947-1.012) 0.915 (0.811-1.032) 1 1.180 (0.972-1.433) 0.981 (0.818-1.177)

0.214 0.149 1 0.094 0.838

1 0.940 (0.740-1.194) 1.008 (0.992-1.025) 1.038 (0.871-1.237) 1 1.056 (0.826-1.350) 1.188 (0.887-1.592)

1 0.611 0.315 0.680 1 0.664 0.248

1 1.111 (0.860-1.435) 1.014 (0.996-1.031) 1.105 (0.916-1.333) 1 1.045 (0.805-1.357) 1.360 (0.994-1.861)

1 0.419 0.129 0.295 1 0.741 0.054

1 1.117 (0.848-1.470) 1.014 (0.995-1.032) 1.172 (0.963-1.427) 1 0.980 (0.742-1.294) 1.442 (1.039-2.001)

1 0.432 0.144 0.114 1 0.886 0.029

RI

PT

OR (95%CI)

SC

AM (year) Category of AM 15-18 years <15 years >18 years Menopause status Pre-menopause Post-menopause ANM (year) Category of ANM 45-55 years <45 years >55 years

Model II

NU

Variable

PT E

D

MA

Model I adjusted for age and BMI. Model II adjusted for age, BMI, education, duration of diabetes, drinking, smoking, and antidiabetic treatment. Model III adjusted for age, BMI, education, duration of diabetes, drinking, smoking, antidiabetic treatment, familial history of diabetes, and physical activity.

Discussion

CE

In the present study, the novel finding indicated ANM is significantly associated with adverse glycemic control of T2DM independent of age, BMI, education, duration of

AC

diabetes, drinking, smoking, antidiabetic treatment, familial history of diabetes and physical activity. Patients with an early ANM (<45 years) or a late ANM (>55 years) had an adverse glycemic control, relative to patients with normal ANM (45-55 years). We also observed that early AM is significantly associated with adverse glycemic control of T2DM independent of age and BMI whereas the association presented marginal P value after adjustment for other confounding factors in model II and model III. And further replication should be warranted in a large sample size of population.

17

ACCEPTED MANUSCRIPT In this present study, results indicated that AM was negatively associated with glycemic control assessed by indices of FPG, HbA1c and ‘FPG & HbA1c’ with adjustment of age and BMI. Previous study reported that among women not taking diabetic medications, earlier menarche (<11 years compared with 13-14 years) was associated with higher level of HbA1c, but not associated with FPG 28. Conversely, a

HbA1c level among groups categorized by AM

29

PT

study consisting of type 1 diabetes patients did not observed significant difference in , and this diversity to our results

RI

may be due to the difference of types of diabetes. Previous studies about AM on glycemic control among T2DM patients were not found, but several research

SC

indicated that early AM could lead to negative consequence in later life: such as high risk of diabetes30; death from all causes31; total number of metabolic syndrome

NU

components32 and cardiovascular disease events33.

MA

Among individuals recruited for treatment studies for obesity co-occurring with binge eating disorder, after adjustment for age, no significant difference in FPG and HbA1c

D

indices was found between premenopausal and postmenopausal women34. The results

PT E

of the Korean National Health and Nutrition Examination Survey of 2001 indicated that among postmenopausal women without hormone replacement therapy, the age-adjusted odds ratio showed no significance for high fasting glucose (6.1 mmol/L)

CE

compared with premenopausal women35. In this study, postmenopausal women with T2DM had significantly low risk of uncontrolled FPG and uncontrolled HbA1c as

AC

well as poor glycemic control after adjustment for confounding factors in model II, though postmenopausal women had a relatively higher age and longer duration of T2DM. Some studies have reported that postmenopausal status was associated with high risk of diabetes 17, in contrast, in our study, postmenopausal T2DM patients have a lower glycemic level (FPG and HbA1c) compared with premenopausal T2DM patients. A cross-sectional analysis reported that age at menopause was associated with T2DM risk in the U-shaped relationship 36. Similarly, patients with T2DM in this study presented a U-shaped relationship about glycemic control which documented that abnormal ANM was an unfavorable factor to glycemic control in T2DM patients. 18

ACCEPTED MANUSCRIPT

HbA1c reflects the average blood glucose level for the most recent three to four months and is considered an important predictive index of T2DM prognostic events37. In this study, HbA1c and FPG were both assessed as a combined index to help evaluate glycemic status synthetically. Information about AM was collected

PT

retrospectively by self-reported in the present study, which meant that misclassification may have existed. Nevertheless, previous studies have shown that

RI

the actual recorded mean age at menarche is quite consistent with that recalled 30 years later 38. Ascertaining a history of menarche and menopause may help to identify

SC

T2DM women predisposed to increased risk of worse glycemic control. A simple question regarding AM and menopause status as well as ANM in the primary clinical

NU

setting could be a useful screening tool to identify T2DM women who need intervention to carefully manage glucose and prevent complication. Further studies

MA

are needed to explore the mechanism of age at menarche and age at menopause on

D

glycemic control among T2DM women.

PT E

The strength of the present study is its novelty of combined assessment of glycemic control, the large number of T2DM patients, and the ability to adjust for multiple confounders. In addition, patients were selected with natural menopause as well as

CE

menopause occurred before diabetes so as to eliminated the influence of diabetes on menopause. There are several limitations of the current study. Firstly, data of female

AC

history were based on a self-administered questionnaire. Secondly, data of anthropometric measurements at patients’ puberty were not collected.

Conclusion Early age at menarche was associated with worse glycemic control as well as early and late age at natural menopause. Ascertaining the age at menarche and natural menopause in T2DM women may help to identify women predisposed to worse glycemic control. 19

ACCEPTED MANUSCRIPT Contributors Ming Wu and Chong Shen designed the program. Ming Wu, Yu Qin, Xiaoqun Pan, Jian Su, Wencong Du, Enchun Pan, Qin Zhang, Ning Zhang and Hongyan Sheng were responsible for data collection. Jinbo Wen analyzed data and drafted the manuscript. Ming Wu, Yu Qin, Chunlan Liu, Hui Wang and Chong Shen revised the manuscript.

PT

All authors approved the version published.

RI

Acknowledgement

The authors thank the field workers for their contribution to the study and the

SC

participants for their cooperation. This work was supported by grants from Jiangsu Province Medical Innovation Team Program (K201105), Jiangsu Provincial Fourth

NU

“333 Project” (BRA2013107) and the Priority Academic Program for the Development of Jiangsu Higher Education Institutions (Public Health and Preventive

D

AC

CE

PT E

Declarations of interest None.

MA

Medicine).

20

ACCEPTED MANUSCRIPT References 1.

Xu Y, Wang L, He J, et al. Prevalence and control of diabetes in Chinese adults. Jama. 2013;310(9):

948-959. 2.

Li YC, Liu XT, Hu N, Jiang Y, Zhao WH. [Disease burden on diabetes in China, 2010]. Zhonghua liu

xing bing xue za zhi = Zhonghua liuxingbingxue zazhi. 2013;34(1): 33-36. 3.

Danaei G, Lawes CM, Vander Hoorn S, Murray CJ, Ezzati M. Global and regional mortality from

ischaemic heart disease and stroke attributable to higher-than-optimum blood glucose concentration: comparative risk assessment. Lancet. 2006;368(9548): 1651-1659. Emerging Risk Factors C, Sarwar N, Gao P, et al. Diabetes mellitus, fasting blood glucose

PT

4.

concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet. 2010;375(9733): 2215-2222.

Gu K, Cowie CC, Harris MI. Mortality in adults with and without diabetes in a national cohort of

RI

5.

the U.S. population, 1971-1993. Diabetes Care. 1998;21(7): 1138-1145.

Schunk M, Reitmeir P, Schipf S, et al. Health-related quality of life in subjects with and without

SC

6.

Type 2 diabetes: pooled analysis of five population-based surveys in Germany. Diabetic Medicine. 2012;29(5): 646-653.

Danforth KN, Townsend MK, Curhan GC, Resnick NM, Grodstein F. Type 2 diabetes mellitus and

NU

7.

risk of stress, urge and mixed urinary incontinence. The Journal of urology. 2009;181(1): 193-197. 8.

Ebbesen MH, Hannestad YS, Midthjell K, Hunskaar S. Diabetes and urinary incontinence -

9.

MA

prevalence data from Norway. Acta obstetricia et gynecologica Scandinavica. 2007;86(10): 1256-1262. Bang WJ, Lee JY, Koo KC, Hah YS, Lee DH, Cho KS. Is Type-2 Diabetes Mellitus Associated With

Overactive Bladder Symptoms in Men With Lower Urinary Tract Symptoms? Urology. 2014;84(3): 670-674.

D

10. Stratton IM, Adler AI, Neil HA, et al. Association of glycaemia with macrovascular and

PT E

microvascular complications of type 2 diabetes (UKPDS 35): prospective observational study. BMJ (Clinical research ed). 2000;321(7258): 405-412. 11. Qiu CS, Chen HJ, Wen JP, et al. Associations Between Age at Menarche and Menopause With Cardiovascular Disease, Diabetes, and Osteoporosis in Chinese Women. J Clin Endocr Metab.

CE

2013;98(4): 1612-1621.

12. Prentice P, Viner RM. Pubertal timing and adult obesity and cardiometabolic risk in women and men: a systematic review and meta-analysis. Int J Obes (Lond). 2013;37(8): 1036-1043.

AC

13. Saquib N, Kritz-Silverstein D, Barrett-Connor E. Age at menarche, abnormal glucose tolerance and type 2 diabetes mellitus: The Rancho Bernardo Study. Climacteric : the journal of the International Menopause Society. 2005;8(1): 76-82. 14. He CY, Zhang CL, Hunter DJ, et al. Age at Menarche and Risk of Type 2 Diabetes: Results From 2 Large Prospective Cohort Studies. American journal of epidemiology. 2010;171(3): 334-344. 15. Harlow SD, Gass M, Hall JE, et al. Executive summary of the Stages of Reproductive Aging Workshop + 10: addressing the unfinished agenda of staging reproductive aging. Menopause. 2012;19(4): 387-395. 16. Kim C. Does menopause increase diabetes risk? Strategies for diabetes prevention in midlife women. Women's health (London, England). 2012;8(2): 155-167. 17. Heianza Y, Arase Y, Kodama S, et al. Effect of Postmenopausal Status and Age at Menopause on Type 2 Diabetes and Prediabetes in Japanese Individuals: Toranomon Hospital Health Management Center Study 17 (TOPICS 17). Diabetes Care. 2013;36(12): 4007-4014. 21

ACCEPTED MANUSCRIPT 18. Kim C, Edelstein SL, Crandall JP, et al. Menopause and risk of diabetes in the Diabetes Prevention Program. Menopause. 2011;18(8): 857-868. 19. Appiah D, Winters SJ, Hornung CA. Bilateral Oophorectomy and the Risk of Incident Diabetes in Postmenopausal Women. Diabetes Care. 2014;37(3): 725-733. 20. Brand JS, van der Schouw YT, Onland-Moret NC, et al. Age at Menopause, Reproductive Life Span, and Type 2 Diabetes Risk. Diabetes Care. 2013;36(4): 1012-1019. 21. Kalyani RR, Franco M, Dobs AS, et al. The association of endogenous sex hormones, adiposity, and insulin resistance with incident diabetes in postmenopausal women. The Journal of clinical endocrinology and metabolism. 2009;94(11): 4127-4135.

PT

22. Ding EL, Song Y, Manson JE, Rifai N, Buring JE, Liu S. Plasma sex steroid hormones and risk of developing type 2 diabetes in women: a prospective study. Diabetologia. 2007;50(10): 2076-2084.

RI

23. Miao DD, Pan EC, Zhang Q, Sun ZM, Qin Y, Wu M. Development and Validation of a Model for Predicting Diabetic Nephropathy in Chinese People. Biomedical and Environmental Sciences.

SC

2017;30(2): 106-+.

24. American Diabetes A. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33 Suppl 1: S62-69. research and reviews. 2016;32(5): 442-458.

NU

25. Weng J, Ji L, Jia W, et al. Standards of care for type 2 diabetes in China. Diabetes/metabolism 26. Wu M, Wen J, Qin Y, et al. Familial History of Diabetes is Associated with Poor Glycaemic Control

MA

in Type 2 Diabetics: A Cross-sectional Study. Sci Rep. 2017;7(1): 1432. 27. Nelson HD. Menopause. Lancet. 2008;371(9614): 760-770. 28. Mueller NT, Duncan BB, Barreto SM, et al. Earlier age at menarche is associated with higher diabetes risk and cardiometabolic disease risk factors in Brazilian adults: Brazilian Longitudinal Study

D

of Adult Health (ELSA-Brasil). Cardiovascular diabetology. 2014;13. 29. Gomes MB, Negrato CA, Calliari LEP, Grp BTDS. Early age at menarche: A risk factor for

PT E

overweight or obesity in patients with type 1 diabetes living in urban areas? Diabetes research and clinical practice. 2015;107(1): 23-30.

30. Conway BN, Shu XO, Zhang XL, et al. Age at Menarche, the Leg Length to Sitting Height Ratio, and Risk of Diabetes in Middle-Aged and Elderly Chinese Men and Women. Plos One. 2012;7(3).

CE

31. Charalampopoulos D, McLoughlin A, Elks CE, Ong KK. Age at Menarche and Risks of All-Cause and Cardiovascular Death: A Systematic Review and Meta-Analysis. American journal of epidemiology. 2014;180(1): 29-40.

AC

32. Feng Y, Hong XM, Wilker E, et al. Effects of age at menarche, reproductive years, and menopause on metabolic risk factors for cardiovascular diseases. Atherosclerosis. 2008;196(2): 590-597. 33. Lakshman R, Forouhi NG, Sharp SJ, et al. Early Age at Menarche Associated with Cardiovascular Disease and Mortality. J Clin Endocr Metab. 2009;94(12): 4953-4960. 34. Udo T, McKee SA, White MA, Masheb RM, Barnes RD, Grilo CM. Menopause and metabolic syndrome in obese individuals with binge eating disorder. Eating behaviors. 2014;15(2): 182-185. 35. Kim HM, Park J, Ryu SY, Kim J. The effect of menopause on the metabolic syndrome among Korean women. Diabetes Care. 2007;30(3): 701-706. 36. Yang A, Liu S, Cheng N, et al. Reproductive factors and risk of type 2 diabetes in an occupational cohort of Chinese women. Journal of diabetes and its complications. 2016;30(7): 1217-1222. 37. Goldstein DE, Little RR, Lorenz RA, et al. Tests of glycemia in diabetes. Diabetes Care. 2004;27(7): 1761-1773. 22

ACCEPTED MANUSCRIPT 38. Must A, Phillips SM, Naumova EN, et al. Recall of early menstrual history and menarcheal body size: after 30 years, how well do women remember? American journal of epidemiology. 2002;155(7):

AC

CE

PT E

D

MA

NU

SC

RI

PT

672-679.

23