Association of Uterine Leiomyoma with Type 2 Diabetes Mellitus in Young Women: A Population-Based Cohort Study
Article information
Abstract
Background
We investigated the association between uterine leiomyoma (UL) and incident type 2 diabetes mellitus (T2DM) in young women.
Methods
A nationwide population-based cohort study of 2,541,550 women aged between 20 and 40 years was performed using the National Health Information Database. Cox proportional hazards models were used to analyze the risk of incident T2DM according to the presence of UL and myomectomy.
Results
The mean age was 29.70 years, and mean body mass index was 21.31 kg/m2. Among 2,541,550 participants, 18,375 (0.72%) women had UL. During a median 7.45 years of follow-up, 23,829 women (0.94%) were diagnosed with T2DM. The incidence of T2DM in women with UL (1.805/1,000 person-years) was higher than in those without UL (1.289/1,000 person-years). Compared with women without UL, women with UL had a higher risk of incident T2DM (hazard ratio, 1.216; 95% confidence interval [CI], 1.071 to 1.382). Women with UL who did not undergo myomectomy had a 1.505 times (95% CI, 1.297 to 1.748) higher risk for incident T2DM than women without UL. However, women with UL who underwent myomectomy did not have increased risk for incident T2DM.
Conclusion
Young women with UL were associated with a high risk of incident T2DM. In addition, myomectomy seemed to attenuate the risk for incident T2DM in young women with UL.
Highlights
• Young women with UL were associated with a high risk of incident T2DM.
• Myomectomy seemed to attenuate the risk for incident T2DM in young women with UL.
• T2DM screening and prevention may reduce the risk of T2DM in young women with UL.
INTRODUCTION
Uterine leiomyoma (UL) is one of the most common benign tumors of the uterus in reproductive age women [1,2]. UL is often asymptomatic and rarely becomes malignant; however, it may cause pelvic pain, dysmenorrhea, excessive uterine bleeding, and infertility in women of reproductive age. Old age, younger age at menarche, and premenopausal state are risk factors for developing UL, as it is linked to longer duration of exposure to ovarian hormones [3,4].
The prevalence and incidence of UL has recently been gradually increasing in young women [5]. Many studies have reported that UL was associated with metabolic disorders [6-9]. A study found that metabolic syndrome and each of their components were associated with an increased prevalence of UL [6]. Insulin resistance was suggested as one of the possible pathophysiological explanations for UL development [10,11]. Hyperinsulinemia induced by insulin resistance can upregulate serum insulin-like growth factor-1 and epidermal growth factor levels [12-14]. This can affect the development of UL by enhancing either ovarian hormone secretion or myometrial smooth muscle cell proliferation, or both. Since insulin resistance is associated with UL, women with UL may be at increased risk of type 2 diabetes mellitus (T2DM), which has insulin resistance as its main pathophysiology. However, no studies have investigated whether UL have associated with incident T2DM, especially in young women. The aim of this study was to evaluate the association between UL and incident T2DM in young women using a nationwide population-based cohort.
METHODS
Data source
We conducted a nationwide population-based cohort study using claims and health checkup data from the National Health Insurance Service-National Sample Cohort (NHIS-NSC). The NHIS is a single-payer health care system for all Korean residents managed by the Korean government, covering more than 98% of the Korean population. The claims data encompasses sociodemographic variables, diagnosis statements defined by the International Classification of Diseases, 10th revision (ICD-10), prescriptions, and hospital visits dates. Our study protocol was approved by the Institutional Review Board (IRB) of Kangbuk Samsung hospital (KBSMC-07-034). Informed consent was waived by IRB as the data set provided from NHIS is deidentified to protect privacy.
Study design and participants
We selected 2,755,790 women aged between 20 and 40 years at baseline who underwent at least one health checkup between 2009 and 2012. The time point at the first health checkup for each patient between 2009 and 2012 was considered the index date. We excluded those who were previously diagnosed with T2DM (n=28,136), those who had hysterectomy during the study period (n=18,485), and those with missing variable for analysis (n=166,012). To overcome bias, we also excluded those who those diagnosed with T2DM within the first year of entry (1-year lag-period; n=1,607). Finally, a total of 2,541,550 patients were eligible for inclusion and were followed until December 31, 2018.
Measurements
Blood pressure (both systolic and diastolic values) was measured by a trained clinician while the participants were seated. Blood samples were obtained after at least 8 hours of fasting. The levels of glucose, total cholesterol, triglycerides (TG), high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and creatinine were measured. Estimated glomerular filtration rate (eGFR) was calculated using the equation from the Modification of Diet in Renal Disease study: eGFR=186.3×serum creatinine−1.154×age−0.203×0.742 (for women).
Information about lifestyle factors, including smoking status, alcohol consumption, and regular exercise was obtained using a standardized self-assessment questionnaire. Drinking status was divided into three categories: nondrinker, mild drinker, and heavy drinker. Heavy drinker was defined as drinking more than 30 g of alcohol per day and mild drinker was defined as drinking less than 30 g of alcohol per day. Regular exercise was defined as more than 30 minutes of moderate intensity activity performed at least five times per week, or more than 20 minutes of high intensity activity performed at least three times per week. Low income was defined as being within the lowest 20% of the total population-based on monthly income.
Definitions
UL was identified in case of female patient admittance with ICD-10 code D25 at least once or diagnosis with ICD-10 codes D25 upon outpatient clinic visit at least twice within one year from the index date. Myomectomy was defined as an ICD-10 code for myomectomy (QZ961, R4120-4126, R4130, R4141, R4145, R4146, RZ564-566, HA636, HA621, M6644) logged within 1 year of the index date.
Hypertension was defined as blood pressure ≥140/90 mm Hg or having been prescribed anti-hypertensive drugs under ICD-10 codes I10–13, I15. Dyslipidemia was defined using ICD-10 codes E78 and at least one record per year for a prescription for a lipid-lowering agent or by a total cholesterol level ≥240 mg/dL. Impaired fasting glucose (IFG) was defined as fasting plasma glucose (FPG) 100 to 125 mg/dL, and chronic kidney disease (CKD) was defined as eGFR <60 mL/min/1.73 m2. T2DM was defined as having FPG ≥126 mg/dL or having been prescribed anti-diabetic drugs under ICD-10 codes E11–E14.
Study outcomes and follow-up
The primary outcome was the incidence of T2DM. The study population was followed from the index date to the date of incident T2DM, or until December 31, 2018 whichever came first. The median follow-up duration was 7.45 years.
Statistical analyses
Continuous variables are presented as mean±standard deviation or median (interquartile range), while categorical variables are presented as numbers (%). Baseline clinical characteristics were compared between groups using Student’s t-test for continuous variables and chi-square test for categorical variables. Incidence rates are presented as the number of events occurring per 1,000 person-years. Cox proportional hazards regression model was used to estimate hazard ratio (HR) and 95% confidence interval (CI) for T2DM. Model 1 was adjusted for age, smoking status, alcohol consumption and regular exercise. Model 2 was additionally adjusted for hypertension, dyslipidemia, and body mass index (BMI). The cumulative incidence of T2DM according to the presence of UL and myomectomy was estimated using the Kaplan-Meier method after adjusted for age, smoking status, alcohol consumption, regular exercise, hypertension, dyslipidemia, BMI, and eGFR. Subgroup analyses were performed according to age, BMI, smoking status, alcohol consumption, regular exercise, and having hypertension, dyslipidemia, and CKD. All data analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA). A P<0.05 was considered significant.
RESULTS
Baseline characteristics of patients are presented in Table 1. A total of 2,541,550 women were eligible, of whom 18,375 (0.72%) were diagnosed with UL. The mean age was 29.70 years, and mean BMI was 21.31 kg/m2. Women with UL were older, less likely to be current smokers and heavy drinkers, less likely to have low household income, and more likely to have hypertension, dyslipidemia, and IFG than those without UL. In addition, the BMI, blood pressure (systolic and diastolic), FPG, total cholesterol, TG, and LDL cholesterol were elevated, while HDL cholesterol was decreased in women with UL. There were no significant differences between women with UL and those without UL with respect to the percentages of patients who undertook regular exercise and who had CKD.
Baseline characteristics of participants according to the presence of uterine leiomyoma
Over a median 7.45 years of follow-up, 23,829 women (0.94%) were diagnosed with T2DM (Table 2). The incidence rate of T2DM in women with UL (1.805/1,000 person-years) was higher than that in those without UL (1.289/1,000 person-years). The cumulative incidence of T2DM in women with UL was higher than that in those without UL (P<0.001) (Fig. 1A). The unadjusted HR for incident T2DM in women with UL was 1.403 (95% CI, 1.236 to 1.593). Although the HR for incident T2DM was slightly decreased after adjustment for age, smoking status, alcohol consumption, and regular exercise (model 1), women with UL still showed an elevated HR for incident T2DM (HR, 1.195; 95% CI, 1.053 to 1.358). In multivariable-adjusted model (model 2), the HR for incident T2DM in women with UL was 1.216 (95% CI, 1.071 to 1.382).
The risk of type 2 diabetes mellitus according to the presence of uterine leiomyoma and myomectomy
The cumulative incidence of type 2 diabetes mellitus. (A) Stratified according to presence of uterine leiomyoma adjusted for age, smoking status, alcohol consumption, regular exercise, hypertension, dyslipidemia, body mass index, and estimated glomerular filtration rate (P<0.001). (B) Stratified according to myomectomy adjusted for age, smoking status, alcohol consumption, regular exercise, hypertension, dyslipidemia, body mass index, and estimated glomerular filtration rate (P<0.001).
Women with UL who underwent myomectomy were older and had higher BMI than those with UL who did not undergo myomectomy (Supplementary Table 1). There was no significant difference in other baseline characteristics between women with UL who underwent myomectomy and those with UL who did not undergo myomectomy. Compared with women without UL, women with UL who did not undergo myomectomy had a 1.505 times (95% CI, 1.297 to 1.748) higher risk for incident T2DM, as opposed to no increased risk in those with UL who underwent myomectomy (Table 2). Women with UL who did not undergo myomectomy still had an elevated HR for incident T2DM (HR, 1.287; 95% CI, 1.108 to 1.495) after adjustment for age, smoking status, alcohol consumption and regular exercise (model 1). In the multivariable-adjusted model (model 2), the HR for incident T2DM in women with UL who did not undergo myomectomy was 1.328 (95% CI, 1.143 to 1.542). However, the risk for T2DM was not increased in women with UL who underwent myomectomy after adjustment for aforementioned factors (model 1 and model 2). The cumulative incidence of T2DM in women with UL who did not undergo myomectomy was higher than that in those with UL who underwent myomectomy (P<0.001) (Fig. 1B).
Subgroup analyses stratified by age, BMI, smoking status, alcohol consumption, regular exercise, hypertension, dyslipidemia, and CKD were performed (Table 3). Women with UL showed a higher risk for incident T2DM than women without UL in almost all subgroups, except in those with BMI ≥25 kg/m2 or those with CKD.
Subgroup analysis for the risk of type 2 diabetes mellitus according to the presence of uterine leiomyoma
DISCUSSION
This nationwide population-based cohort study demonstrated that young women with UL were associated with a high risk of incident T2DM. Incidence rate of T2DM in women with UL was higher than that in women without UL. Interestingly, women with UL who underwent myomectomy did not have an increased risk for incident T2DM compared with those without UL. However, in women with UL who did not undergo myomectomy, the risk for incident T2DM was higher than that in those without UL. To our knowledge, this was the first study to evaluate the association between UL and incident T2DM in young women using a nationwide population-based cohort.
Many studies have shown that UL was associated with obesity and metabolic disorders [2,6-9]. An observational case-control study including 71 women of childbearing age diagnosed with UL and 145 women as control group, preperitoneal fat thickness showed an independent association with the presence of UL [7]. Another study, including 89 patients with UL and 81 healthy women without UL, showed that increased body fat (especially abdominal visceral fat) is associated with an increased risk of UL [8]. Recently, a meta-analysis examining 24 studies found a positive association between obesity and the risk/prevalence of UL (odds ratio [OR], 1.19; 95% CI, 1.09 to 1.29) [2]. Takeda et al. [9] have reported that BMI, blood pressure, TG, and FPG were significantly higher in 213 women with UL compared to those in 159 women without UL and the risk of UL was correlated with increased metabolic component levels. Other study that enrolled 1,230 parous premenopausal women reported that women with UL had higher waist circumferences, body fat, blood pressure, and LDL cholesterol than women without UL [6]. In our study, women with UL had poor metabolic profiles, such as higher blood pressure, FPG, total cholesterol, TG, LDL cholesterol, and lower HDL cholesterol compared with women without UL.
However, few studies have evaluated the association between UL and T2DM. Aforementioned study enrolled 1,230 parous premenopausal women revealed that hyperglycemia is significantly associated with an increased risk of UL (OR, 1.45; 95% CI, 1.10 to 1.89) [6]. In a Taiwanese population-based cohort study, metformin use was associated with a low risk of UL in 21,996 women with new-onset T2DM (HR, 0.467; 95% CI, 0.387 to 0.564) [15]. On the other hand, a nested case-control study enrolling 3,789 participants reported a protective association between UL and T2DM, but it was a cross-sectional design that included relatively old women (mean age, 47 years in UL group; 44 years in control group) [16]. In our study, women with UL had 21.6% higher risk for incident T2DM compared with those without UL. The strength of our study was that it was a longitudinal design over a median 7.45 years of follow-up and more than 2 millions women were enrolled, which made it much larger than in the aforementioned studies.
The pathophysiology of how UL affects the development of T2DM is unclear. Along with estrogen, progesterone and progesterone receptors are reported to play a key role in the development and growth of UL [17]. Increased progesterone level is associated with reduced insulin sensitivity, which can lead to development of T2DM [18,19]. Although insulin sensitivity may be reduced by the effect of progesterone in women with UL, it is difficult to determine whether there is a casual relationship between the presence of UL and development of T2DM. As there is a possibility that reduced insulin sensitivity is a common risk factor for the development of T2DM and UL. However, an animal study showing improved glucose homeostasis in progesterone receptor knockout mice supports the adverse effect of progesterone on glucose tolerance and insulin release [20]. In our study, women with UL who underwent myomectomy did not have an increased risk for incident T2DM compared with those without UL, whereas women with UL who did not undergo myomectomy still had a high risk for incident T2DM. A possible mechanism for the reduction of incident T2DM risk associated with myomectomy may be due to the reduction of progesterone receptor expression through the mechanical removal of UL.
In subgroup analysis stratified by BMI, women with BMI ≥25 kg/m2 revealed no significant difference in the risk of incident T2DM between women with UL and without UL. Similarly, there was no difference in the risk of incident T2DM between women with UL and without UL in women with CKD. This might be because obesity and CKD have been already strong risk factors for T2DM [21-23]. Relatively higher incidence rate of T2DM in women with obesity or CKD than those without obesity or CKD support this hypothesis (Table 3). In addition, when we analyzed in women without obesity or CKD, the results were similar with main study (Supplementary Table 2). Women aged more than 30 years tended to have no significant difference in the risk of incident T2DM between women with UL and without UL. Old age is also very strong risk factor for T2DM [24,25]. Because women more than 30 years had already higher incidence rate of T2DM than women less than 30 years, there may have been no difference in the risk of incident T2DM between women with UL and without UL in women more than 30 years. Meanwhile, UL tended to associated with an increased risk for incident T2DM even in women without hypertension, dyslipidemia, and CKD. Because one of the major risk factors for hypertension, dyslipidemia, and CKD is insulin resistance [26,27], we could assume that UL itself increases the risk of T2DM through mechanisms other than insulin resistance. Further prospective study is warranted to determine the association between UL and incident T2DM.
Several limitations should be considered when interpreting the results of this study. First, since UL is mostly asymptomatic in early stages, women who have asymptomatic UL may have been undiagnosed. Second, information about the size and number of ULs was not available due to the database characteristics. Third, gynecologic information, including parity, presence of polycystic ovarian syndrome, or hormonal status, were also inaccessible; these factors can affect the development of T2DM, and their absence lowers the generalizability of our results. Fourth, we could not get the information on various insulin resistance indices due to limitation of database. Finally, because this study was conducted using the Korean nationwide database, the results should be interpreted with caution when applied to other ethnicities. This study was nonetheless valuable because it was the first large-scale, longitudinal study to evaluate the association between UL and the risk of incident T2DM in young women.
In conclusion, young women with UL were associated with a high risk of incident T2DM. Furthermore, myomectomy seemed to attenuate the risk for incident T2DM in young women with UL. This study suggests that T2DM screening and prevention might be helpful to reduce the risk of incident T2DM in young women with UL.
SUPPLEMENTARY MATERIALS
Supplementary materials related to this article can be found online at https://doi.org/10.4093/dmj.2023.0444.
Baseline characteristics of women with uterine leiomyoma according to myomectomy
The risk of type 2 diabetes mellitus according to the presence of uterine leiomyoma and myomectomy in women without obesity or chronic kidney disease
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conception or design: J.H.S., K.S.K., C.Y.P.
Acquisition, analysis, or interpretation of data: all authors.
Drafting the work or revising: J.H.S., K.S.K., C.Y.P.
Final approval of the manuscript: all authors.
FUNDING
None
Acknowledgements
None