INTRODUCTION
The menstrual cycle mainly consists of the ovarian and uterine cycles, which are regulated by hormones from the hypothalamic-pituitary-ovarian (HPO) axis [1]. Menstrual cycle disturbances include disturbances of regularity, quantity, frequency, and duration [2]. Approximately 14.2% to 27% of women of reproductive age worldwide experience menstrual cycle irregularities [3-5]. This irregularity is often the result of various benign factors, including weight change, excessive exercise, dietary habits, stress, medication, and smoking, but may also reflect pathologic conditions such as polycystic ovary syndrome (PCOS) and other hormonal imbalances or reproductive disorders [1,6,7]. Although menstrual cycle irregularity is a multifactorial phenomenon, the disturbances are mostly dependent on the HPO axis, which is also partly affected by liver [5,8-11].
The liver and reproductive system are known to have a bidirectional relationship, whereby the liver plays an important role in gonadal steroid hormone metabolism, its transport to tissues, and other metabolic pathways that could affect reproductive health in general. Gonadal steroid hormone signaling influences the pathogenesis of hepatic disorders [8]. Hence, decreased liver function can induce reproductive dysfunction, which can present as menstrual cycle irregularities [12]. Previous studies have shown that non-alcoholic fatty liver disease (NAFLD) is associated with PCOS, which often presents with menstrual irregularity, depending on insulin resistance and hyperinsulinemia [12-17]. Furthermore, several studies have shown the normalization of reproductive function, including the menstrual cycle, after liver transplantation in patients with liver diseases [18,19]. It is possible to presume that patients with compromised liver function or NAFLD may have a higher probability of experiencing menstrual cycle irregularity.
Similarly, certain previous studies have investigated the association between the liver and the reproductive system. However, these studies have mainly focused on selected populations, including patients with specific hepatic disorders or post-liver transplantation patients [12,13,17-20]. Consequently, evidence from large population-based studies is currently limited. Liver function markers could be considered convenient tools for monitoring liver function in the general population during regular health checkups.
Therefore, this study aimed to investigate the association between liver function markers and menstrual cycle irregularities in a premenopausal Korean female population using nationwide data.
METHODS
1. Data Source
This nationwide study analyzed data from the 5th Korea National Health and Nutrition Examination Survey (KNHANES) conducted from 2010 to 2011, during which relevant data were collected. The KNHANES is a nationwide survey carried out by the Korean Ministry of Health and Welfare and the Korean Centers for Disease Control and Prevention since 1998. The sample surveys are extracted and conducted annually from January to December and consist of health, nutrition, and examination surveys.
All participants signed an informed consent form prior to the survey, which was reviewed and approved by the Institutional Review Board (IRB) of the Korean Centers for Disease Control and Prevention (IRB approval no., 2010-02CON-21-C and 2011-02CON-06-C).
2. Study Population
Among the total 17,476 participants in the KNHANES 2010–2011, 14,431 were excluded (Figure 1). The exclusion criteria consisted of men (n=7,982), postmenopausal women (n=4,082), pregnancy (n=64), age (under 19 years or over 60 years, n=1,879), history of bilateral salpingo-oophorectomy operation and missing menstrual data (n=12), absence of liver enzymes data (n=94), hepatitis B, C, liver cirrhosis, or hepatic cancer (n=16), and missing covariates (n=302). Finally, the study included 3,045 premenopausal women.
3. Liver Function Markers
Liver function markers include biochemical markers representing the degree of liver damage, including serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyl transferase (GGT) [21]. Fatty liver index (FLI) is a noninvasive method for measuring hepatic steatosis and assists in the selection of candidates who are recommended to undergo hepatobiliary ultrasound evaluation [22]. FLI was calculated using the following equation [22]:
FLI=eL / (1+eL)×100
where L=0.953×loge (triglycerides)+0.139×BMI+0.718×loge (GGT)+0.053×waist circumference–15.745
These markers can be routinely evaluated during health checkups and may be elevated in hepatic parenchymal pathologies. In particular, ALT is an enzyme known to be more specific and predominant in the liver than in other organs compared to AST [21].
4. Reproductive Factors
In this study, data on reproductive factors including menstrual cycle regularity, age at menarche, oral contraceptive use, and history of pregnancy were obtained through a self-administered questionnaire. Participants were asked to record their menstrual cycle characteristics. The question about the presence of a regular menstrual cycle was asked, and those who answered “yes” were considered to have regular menstrual cycles, while those who answered “no” were considered to have irregular cycles. Menstrual cycle irregularity was further defined by inquiring about the inter-menstrual duration, specifically whether it was within or exceeded 3 months.
5. Other Variable Selection
Sociodemographic and lifestyle variables including age, alcohol consumption, smoking, physical activity, residential area, and stress perception were considered as covariates. Data on age was collected using the self-administered questionnaire. Those aged 19 years or older were classified as adults. Alcohol consumption within a month of the survey was classified as drinkers and others as non-drinkers. Smoking data obtained from the self-administered questionnaire was classified as current smokers, ex-smokers, and non-smokers. Using the International Physical Activity Questionnaire, physical activity was defined as moderate-intensity exercise for more than 30 minutes a day, 5 days a week, or vigorous-intensity exercise for more than 20 minutes a day, 3 days a week. As for the residential area, subjects who reside in “dong” were defined as urban residents, and those who live in “eup” or “myun” were defined as rural residents. Stress perception was assessed using the self-administered questionnaire.
Anthropometric measurements, including height (m) and weight (kg), were obtained and body mass index (BMI), was calculated by dividing the weight by the square of the height. Waist circumference was measured during normal exhalation. Blood pressure was measured consecutively 3 times in the right arm using a mercury blood pressure meter after a rest of more than 5 minutes. The average of the second and third measurements was used.
Laboratory findings were obtained from a blood test performed after fasting for at least 8 hours. Fasting blood glucose and serum lipid profile (total cholesterol, high-density lipoprotein [HDL] cholesterol, low-density lipoprotein [LDL] cholesterol, and triglycerides) values were included as research data.
Metabolic syndrome was defined by using the criteria of National Cholesterol Education Program Adult Treatment Panel III as fulfilling any three or more of the following conditions: (1) fasting plasma glucose ≥110 mg/dL; (2) blood pressure ≥130/85 mm Hg; (3) triglycerides ≥150 mg/dL; (4) HDL cholesterol <50 mg/dL for women; and (5) waist circumference ≥88 cm for women [23].
6. Statistical Analysis
The data obtained from the KNHANES were analyzed by applying sampling weights to accurately estimate the characteristics of the study population. The baseline characteristics of the study population are represented either as the mean and standard error of the mean or number of participants and weighted percentage. The liver function markers, ALT, AST, and GGT, were expressed as both continuous and categorical variables and grouped into quartiles according to the measured values. In other published studies, an FLI cutoff value of ≥60 was used to rule out hepatic steatosis, and the same cutoff was applied in this study [22]. The comparative baseline characteristics according to the presence or absence of menstrual regularity were analyzed using the t-test for continuous variables and the chi-square test for categorical variables. Analysis of variance was used to evaluate the variance in the mean values of liver function markers according to menstrual cycle regularity and duration. Analysis of covariance adjusted for age, alcohol consumption, smoking, and history of pregnancy was also performed for the mean FLI values.
To evaluate the association between liver function markers and menstrual cycle irregularity, multivariate logistic regression analysis was performed and P-values were calculated. Hierarchical adjustment was performed where model 1 was a non-adjusted value; model 2 was adjusted for social factors such as age, alcohol consumption, and smoking; model 3 was additionally adjusted for factors related to obesity and metabolic diseases including BMI, waist circumference, LDL cholesterol, HDL cholesterol, triglycerides, and metabolic syndrome; and model 4 was additionally adjusted for history of pregnancy. The same analysis was performed separately for the FLI, and the covariates of age, alcohol consumption, smoking status, and history of pregnancy were adjusted. Stratified subgroup analysis was also performed, and P-values for interactions were calculated. Statistical analysis was performed using SAS ver. 9.3 (SAS Institute Inc., Cary, NC, USA), and the statistical significance level was set at P-value <0.05.
RESULTS
1. Baseline Characteristics
Table 1 shows the baseline characteristics of the study population. Among the 3,045 participants, 2,633 (85.6%) had regular menstrual cycles, and 412 (14.4%) experienced menstrual irregularity. There were no differences between the groups in age (P=0.536), alcohol consumption (P=0.207), smoking (P=0.067), physical activity (P=0.892), residential area (P=0.643), or stress perception (P=0.054).
Individuals with menstrual irregularity had significantly higher BMI (P<0.001), waist circumference (P<0.001), LDL cholesterol (P=0.024), and triglycerides (P=0.030) but lower HDL cholesterol (P=0.012). However, those with menstrual irregularity showed a significantly higher prevalence of metabolic syndrome (P=0.011), whereas reproductive factors were associated with a lower history of pregnancy (P=0.002).
Regarding liver function markers, serum ALT and AST levels were significantly higher in the irregular menstrual cycle group (P<0.001) in the continuous variable form. The FLI was also significantly higher in women with irregular menstrual cycles (P<0.001). However, the serum GGT levels did not differ significantly between the groups (P=0.186).
2. Association between Liver Function Markers and Menstrual Cycle Irregularity
The associations between liver function markers and menstrual cycle irregularity are presented in Table 2 as adjusted odds ratios (ORs) with 95% confidence intervals (CIs). In model 1 where no adjustment was made, the highest quartile of serum ALT level and serum AST level had higher ORs for menstrual cycle irregularity (OR, 2.09; 95% CI, 1.47–2.97; OR, 1.74; 95% CI, 1.24–2.43, respectively). In model 4 where multiple covariates were adjusted, the highest quartile of serum ALT level had significantly higher ORs for menstrual cycle irregularity (adjusted OR, 1.83; 95% CI, 1.26–2.64). A similar association between serum AST level and menstrual cycle irregularity (adjusted OR, 1.67; 95% CI, 1.17–2.39) was also observed.
On the contrary, for serum GGT level, the highest quartile of serum GGT level had higher ORs for menstrual irregularity (OR, 1.53; 95% CI, 1.07–2.19) only in the unadjusted model 1. In the adjusted models, the serum GGT levels were not significantly associated with menstrual cycle irregularities.
3. Association between Fatty Liver Index and Menstrual Cycle Irregularity
The associations between FLI and menstrual cycle irregularity are presented in Table 3 as adjusted ORs with 95% CIs. In model 1 where no adjustment was made, the FLI value over the cutoff value of 60 showed a higher OR for menstrual cycle irregularity (OR, 2.12; 95% CI, 1.33–3.39), and in model 3 where all the covariates were adjusted, it also had significantly higher OR for menstrual cycle irregularity (OR, 2.14; 95% CI, 1.31–3.48). The highest quartile of FLI showed significantly higher ORs for menstrual irregularity in both non-adjusted and adjusted models (OR, 1.60; 95% CI, 1.17–2.18; OR, 1.76; 95% CI, 1.23–2.52, respectively).
4. Subgroup Analysis
A similar result was observed in stratified subgroup analyses. Subgroup analyses were performed according to smoking, alcohol consumption, age, LDL cholesterol, triglycerides, history of pregnancy, and physical activity (Table 4). The association between serum ALT levels and menstrual cycle irregularity did not differ among the subgroups of smoking, alcohol consumption, age, LDL cholesterol, triglycerides, history of pregnancy, and physical activity (P for interactions: 0.113, 0.514, 0.081, 0.079, 0.400, 0.154, and 0.538, respectively).
5. Mean Values of Fatty Liver Index according to Menstrual Cycle Characteristics
The mean of FLI was the highest in those with irregular cycle duration exceeding 3 months (29.3±4.3, P<0.001) as shown in Table 5. A similar trend was observed for the adjusted mean where the highest adjusted mean was seen in the group with irregular cycle duration exceeding 3 months (28.6±4.5, P<0.001).
6. Mean Values of Liver Function Markers according to Menstrual Cycle Characteristics
As shown in Figure 2, the mean serum ALT level was the highest in those with irregular cycle duration exceeding 3 months (25.1±3.5, P=0.001). The highest mean was also observed for serum AST level for irregular cycle duration longer than 3 months (23.9±1.9, P<0.001).
DISCUSSION
In this study, a significant positive association was observed between liver function markers and menstrual cycle irregularities in the premenopausal Korean female population. Compared to the lowest quartile, the highest quartile of serum ALT and AST levels showed almost 1.8- and 1.6-fold greater odds of menstrual cycle irregularities, respectively. The elevation of liver enzymes reflects a decrease in hepatic function, which consequently appears to be associated with menstrual cycle disturbances. Additionally, the group with an FLI value of 60 or higher, which could rule out hepatic steatosis, showed 2.1-fold higher odds of menstrual cycle irregularity, elucidating the association between fatty liver and the menstrual cycle.
Previous studies have investigated more specific and smaller-sized samples. However, some of these findings were too limited to be applied to the general population. They analyzed the relationship between the liver and reproductive system, focusing mainly on the underlying hepatic and hormonal interactions in the male reproductive system or on a subset of patients diagnosed with specific diseases, including vascular liver diseases, as well as patients with severely compromised liver function. Several of these studies have discussed the normalization of the menstrual cycle and other reproductive functions after liver transplantation in patients with compromised liver function [18-20,24-26]. In certain studies, disturbances in the HPO axis and consistently low levels of sex steroid hormones throughout the liver disease progression seem to return to normal levels upon liver transplantation, which ameliorates reproductive functions [18,19].
Moreover, previous studies have described possible hypothetical mechanisms underlying menstrual cycle irregularity in women with various hepatic disorders. The liver modulates the reproductive axis by regulating gonadal steroid hormone metabolism and transport to the tissues [8]. The etiology of the hepatic effect on menstrual cycle disturbances has not been fully elucidated. However, previous studies have found it to be related to hypogonadotropic hypogonadism or hyperandrogenism as seen in PCOS [12,14,27]. The former is reported to be due to liver-dependent dysfunction of hypothalamic-pituitary axis and possible primary gonadal failure [18,19,28]. Hyperandrogenism, may result from portosystemic shunting and hyperinsulinemia in patients with liver diseases, which increases androgen production by theca cells [1,12,18,29]. Additionally, the clearance of androstenedione by the hepatic system is downregulated [12]. The changes in sex hormone-binding globulin (SHBG), which functions as the main binding protein for gonadal hormones and their transport, in patients with hepatic disease remain debatable. Hyperinsulinemia may inhibit the hepatic synthesis of SHBG, which may elevate unbound free gonadal hormone levels [1]. In contrast, some studies have shown that SHBG is elevated in relation to hepatic failure and, in turn, induces decreased metabolic clearance of hormones [1,8,12,18,27]. Recently, several studies on the bidirectional relationship between the liver and reproductive system have shown that gonadal steroid hormone signaling may affect the pathogenesis of liver diseases, where long and irregular menstrual cycles are associated with a higher prevalence of NAFLD [8,13]. Our investigation also reported that those with possible hepatic steatosis could be associated with menstrual cycle irregularity, and those with longer duration of menstrual irregularity had the highest mean values of liver function markers, indicating the potential for a bidirectional interaction.
The first limitation of this study is that liver function was not evaluated using more comprehensive and specific tests. Although the liver function markers included in this study are convenient indicators of hepatic function, more advanced evaluations could be conducted to accurately estimate the hepatic functions of the participants, even though routine monitoring may be limited. Second, menstrual cycle data were obtained using a self-administered questionnaire, which could have created recall bias. Third, due to the lack of information on PCOS, which is associated with menstrual irregularities and NAFLD, we were unable to include it as a confounding factor in the analysis. Finally, considering that this was a cross-sectional study, it was difficult to determine the cause-and-effect relationship over time.
However, this study has definite strengths. This is a large-scale analysis conducted using nationwide data with a large sample size and a multistage probability sampling design, meaning that our findings could potentially be applied to the general population. Additionally, this study used data such as liver function markers, which could be obtained from more accessible tests. It also accounted for and adjusted for potential confounding factors, including age, smoking, alcohol consumption, metabolic syndrome, and history of pregnancy.
In conclusion, liver function markers were positively associated with menstrual cycle irregularities. In clinical settings, women of reproductive age with relatively decreased liver function should be recommended a more thorough and regular follow-up of their reproductive health status.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
