Effects of menarche, parity, primiparous age, and reproductive diseases on uterine fibroids of rural women in China

Objectives: Given the increasing prevalence of hysteromyoma among the Chinese rural women of childbearing age, it is imperative that more attention should be given to researching risk factors that predispose women to this condition so that early preventive measures may be taken. Our study investigates potential risk factors for uterine fibroids such as participants' demographic characteristics (age, ethnicity, education), physiological characteristics (age at menarche, primiparous age), and the occurrence of several different female reproductive diseases (vaginitis, pelvic inflammatory disease). Methods: 2,200 rural women of reproductive age were screened from a cross-sectional study carried out in the Anhui province. Data was collected by questionnaire surveys, gynecological and laboratory examinations and were subjected to univariate and multivariate analysis to evaluate the risk factors for uterine fibroids. Results: Of the 2,200 women, 440 had uterine fibroids. In general, women with endometriosis, ovarian cysts and early menarche were more likely to suffer from hysteromyoma. Factors including higher parity, late age of first childbirth, and a regular menstrual cycle were protective against uterine fibroids. Conclusions: Our findings indicate that parameters such as earlier age of menarche, early primiparous agelower parity and a variety of female reproductive diseases could increase the risk of uterine fibroids. These findings may be used to guide the formulation of effective prevention and intervention protocols to improve the reproductive health of rural women.


Introduction
Uterine fibroids originate from muscle cells in the uterus and are often initially diagnosed as small nodules or polyps. Large fibroids may present as a lower abdominal mass. uterine 3 fibroids or uterine fibroid tumor is the most common female pelvic benign tumor and is classified based on its location relative to the layers of the uterus (as subserous, intramural or submucosa), occurring in 20%-40% in women of reproductive age 1,2 . The incidence of uterine fibroids varies with race and age 3 . Uterine fibroids are steroid dependent 4 . Due to the influence of estrogens and progestogens 5 , uterine fibroids rarely occurs prior to menarche and rarely after the onset of menopause. Women with uterine fibroids may present with increased menstruation, prolonged menstruation, and lower abdominal pain during menstruation. However, many patients remain asymptomatic or possess vague symptoms 6 , leading to late detection and inevitable surgical intervention.
Most patients with large uterine fibroids are often left with the sole treatment option of hysterectomy, which is also the main reason why women in general are hospitalized for gynecological diseases 7, 8 . Studies have suggested that an earlier menarche is a risk factor for uterine fibroids 9 . Early age of menarche has also been associated with other conditions such as type 2 diabetes, angiocardiopathy, metabolic diseases 10,11 , and cancers 12 . The age of menarche is influenced by both genetic and environmental childhood factors 13 , such as nutrition, geographic location, stressful events, light, and the presence of a chronic infectious disease [14][15][16] , thereby supporting the "developmental origins of health and disease" hypothesis.
Thus, exploring the relationship between age of menarche, parity, female reproductive diseases and hysteromyoma is of remarkable clinical and public health importance, especially in the context of the declining age of menarche and increasing prevalence of hysteromyoma.

Setting and sample
This cross-sectional study was conducted between August 2005 and November 2007 in the Anhui province, which is located in the east-central part of China covering an area of 39,600 km 2 and with a population of more than 70 million. At that time, more than 50% of the population resided in rural areas of the province; 17 million women were of childbearing age, and nearly 80% were married 17 . This survey was designed by the Population and Family Planning Commission of Anhui Province and performed by the Department of Public Health of Anhui Medical University.
All participants obtained informed consent and the study was approved by the Anhui Provincial Science and Technology Department and the Research Ethics Committee of Anhui Medical University. Data regarding women's health were collected via questionnaires and clinical gynecological examinations (the latter were considered as the gold standard for diagnosing diseases in this study). The relevant clinical examination was to confirm the presence of female reproductive diseases such as vaginitis and pelvic inflammatory disease, while the questionnaire was to determine the demographic characteristics and previous medical history of the participants. This study is a crosssectional study. Specific methods for participants selecting are described below.
Firstly, the Anhui province was stratified based on administrative region and landforms.
Subsequently, several towns were picked out randomly within the stratification. Secondly, administrative villages situated within a 5 km radius from local town family planning service center were randomly selected. Women who met the inclusion criteria were included in this study. Women who could not recall with certainty their age at first menstruation were excluded to minimize bias in the outcome. A total of 53,652 women were interviewed, of which 52,874 responded to the questionnaires. Due to the limited scientific research funds and human resources, the examination of uterine fibroids was only carried out in three randomly selected villages. An examination specific to ruling in the presence of uterine fibroids was done, however, the location and number of fibroids were not examined in detail. Given that several women that participated in the study from these three towns were also partaking in excessive farm work at home, many did not arrive at the designated examination area after submitting the initial questionnaire. Therefore, women who had provided only questionnaires but without clinical examination data from regions were excluded from the study. A final result of 2,200 samples were finally included in our study.

Questionnaire and date collection
Questionnaire survey was conducted by trained staffs who were instructed to obtain specific information such as the sociodemographic characteristics (age, ethnic, education level, etc.), menstrual history (including menstrual cycle, the days and blood volume during period, menstrual regularity), obstetric history (parity, primiparous age, cesarean section or natural birth and so on), and medical history. Prior to the start of the study, investigators were given permission by the participants to visit their homes and to maintain a record of their home addresses in order to ensure data accuracy and completion (if any problems occurred, we can find them to resolve it). Information of women who arrived at the family planning service station of the villages were recorded.
Residents who failed to show up were visited by investigators.

Gynecological examination and classification
Gynecologists, laboratory technicians, and health workers obtained female reproductive health statuses through gynecological examination and laboratory testing. Initially, subjects were interviewed to determine information such as previous gynecological history, abnormal pelvic symptoms, and hypogastric discomfort or pain. Vaginal swabs and 6 blood samples were obtained for detection of reproductive tract infections. The gynecological diseases that were tested in this study included vaginitis, pelvic inflammatory disease, dysmenorrhea, endometriosis and ovarian cysts. B-type abdominal ultrasound performed by two senior doctors from the radiological department with at least 3 years of experience. The female vagina is a balanced environment with a variety of microorganism's coexistence. When the balance is broken, pathogenic microorganisms can grow and dominate in the vagina 18  children. According to the severity of symptoms, dysmenorrhea was divided into three 7 groups: no (Ref. group), mild, and severe. Most of the remaining dichotomous data were divided into two groups: non-exposed group (no = 0) and exposed group (yes = 1).

Statistical analysis
Two students utilized EpiData3.1 [EpiDataForeningen (EpiData Association), Odense, Denmark] to enter data and SPSS23.0 was used to analyze data separately. We first performed a bivariate analysis with risk estimates for uterine fibroids using odds ratio (OR) with 95% confidence interval (CI) for each predictor. Then, a multivariate analysis using nonconditional multiple logistic regression was applied to jointly assess the risk factors for uterine fibroids, reporting the estimated adjusted OR (OR adj ) with 95% CI. The enter method was used and only predictors with a p-value < 0.05 in the bivariate analysis entered the multivariate model. In all tests, p-value < 0.05 was considered to be statistically significant.

Demographic characteristics as well as reproductive health status
Of the 2,200 women included in this study, 440 had uterine fibroids. The incidence of uterine fibroids was 20%. Participants' demographic characteristics, female physiology, and reproductive diseases are shown in Table 1

The distribution of uterine fibroids among different analyzed variables
There were no statistical differences between the frequency distribution of uterine fibroids in groupings based on ethnicity, occupation, and yearly family income. However, statistical differences were demonstrated between different ages, education level, and marital status (p < 0.05). There was statistical difference in the frequency distribution of uterine fibroids based on menarche, parity, and age of first childbirth (p < 0.001). With regards to the presence of female reproductive diseases, there were significant differences found in the distribution of uterine fibroids based on a history of vaginitis, endometriosis, ovarian cyst, and menstrual disorders (p < 0.001), with the exception of dysmenorrhea and pelvic inflammation. More information is depicted in Table 1.

Risk factors for uterine fibroids
Binary logistics regression analysis with the dependent variable of women with or without fibroids (no = 0, yes = 1) and independent variables of various characteristics revealed that several common sociodemographic factors (ethnicity, occupation, education level, marital status, yearly family income), with the exception of age, were not associated with uterine fibroids (Table 2). In contrast to the age group of those between 26 to 35 years, patients that were more than 35 years old had a higher risk of developing uterine fibroids. Table 3 suggested that AOM (age of menarche), parity, and age of first childbirth were significantly associated with uterine fibroids. Those with early AOM (≤ 12, 9 OR = 1.57, 95%CI: 1.16-2.12) or early primiparous age (≤ 20, OR = 1.36, 95%CI:1.05-

Discussion
Women residing in rural areas are often of lower socioeconomic status and have less access to family planning services compared to cities. It comes as no surprise that rural women are at a higher risk of gynecological diseases. Although the fertility rate of rural women in the Anhui province is high, prevention of gynecological diseases is still insufficient. Related research has confirmed sub-mucosal fibroids are more likely to lead to irregular bleeding and adverse reproductive outcomes in women 23,24 . Prior studies have also illustrated that infertility is associated with uterine fibroids 9,25 , especially at earlier ages 26 . Our study found that 20% of the participant cohort of 2200 cases had fibroids. This is in line with available data 1,2 that states that fibroids are the most common pelvic tumor and is found in 20%-40% of women of child-bearing age. The prevalence of uterine fibroids is also age dependent. In fact, it is equivalent to the effect of estrogen because the secretion of estrogen varies in different age groups. Our research indicated that highest prevalence (68.2%) of fibroids groups was found among women aged [36][37][38][39][40][41][42][43][44][45] years, and of lower prevalence (16.6%) among those aged < 35 years, and were the rarest among (15.2%) women aged ≥ 46 years. Our finding of women between 36-45 years being the most common group to develop uterine fibroids is in agreement with previous studies 27 . This demonstrates that age is a risk factor for uterine fibroids 27,28 . Uterine fibroids are estrogen-dependent, so decreased hormone levels in menopausal women may also explain the lower prevalence (15.2%) of fibroids in those aged above 46 years in the current study. Menstruation represents the periodic shedding of the endometrium that is influenced by various hormones. Early menarche may occur due to higher estrogen levels 29 . Earlier menarche indicates earlier initiation of ovulation and therefore a longer overall duration of exposure to estrogen and progesterone 30 , which could explain our results that early menarche age (≤ 12 years, OR = 1.57, 95%CI: 1.16-2.12) is a risk factor for fibroids, a finding that has also been confirmed by previous studies 9,31 . Furthermore, our results indicate that higher parity was protective against uterine fibroids. There was an increase in uterine fibroid prevalence with decreasing parity 26,32 . One hypothesis for this phenomenon suggests that the protective effect of pregnancy on fibroids occurs during the period of postpartum uterine involution, which resembles uterine remodeling.
This process involves high rates of apoptosis and muscle cell contraction, both of which are necessary for regaining uterine size. Uterine remodeling appears to enhance the selective elimination of abnormal uterine cells, such as cells that form uterine fibroids 33 .
Other reasons behind protective effect of pregnancy is the theoretical reduced exposure to physiological estrogen and progesterone.
Women with symptomatic uterine fibroids have a higher risk of endometriosis 34 , nevertheless, our results suggests that endometriosis is a risk factor (OR = 20.3, 95%CI: 7.7-53.5) for uterine fibroids, which reflects mutual causality.
Earlier studies have revealed that women who have been diagnosed with bacterial vaginosis were more likely to have multiple instead of single fibroids 35 . This is consistent with our study that found a positive association between vaginitis (OR = 4.05, 95%CI: 3.26-5.04) and the risk of developing uterine fibroids.
The symptoms of genital tract infection are usually ignored by women as they are not inherently debilitating or obvious, however, long-term infection can trigger endometrial proliferation, resulting in uterine fibroids 36,37 . Early primiparous age (≤ 20, OR = 1.3, 95%CI: 1.01-1.68) has been associated with a significant increase in the prevalence of uterine fibroids. It is well known that the age of childbirth influences pregnancy outcomes.
The reproductive age with the lowest risk of adverse pregnancy outcome (miscarriage, premature birth and stillbirth) is between 26 to 30 years old. Early ages of first childbirth places the fetus at risk of prematurity and stillbirth; and the mother at risk of gynecological diseases and cancer [38][39][40] . Other studies have shown that miscarriage significantly increases the risk of uterine fibroids 41,42 . Our study also correlates the

Conclusion
In brief, our data indicates that female reproductive diseases and earlier menarche are risk factors for uterine fibroids, while multiple parity provides protection from uterine fibroids. Therefore, in order to reduce the prevalence of female uterine fibroids and improve the reproductive health index, it is necessary to take preventive measures. The relevant medical and health departments advocate for rural women to seek healthcare services in a timely manner to avoid late detection of uterine fibroids. Factors to improve nutrition may also have a role in influencing the onset of menarche include nutrition, a highly modifiable risk factor. The government should exert the 12-year compulsory education policy as soon as possible (every child can enjoy the state compulsory education until high school graduation). Children can learn more knowledge, starting the social work then, and getting married later, which can lest the primiparous age early. The current generation of parents should be responsible to the national two-child policy to avoid having too few children.

Limitations
Significant limiting factors of our study are funding and human resources, which in turn stunted our ability to provide a more detailed classification of the various types of gynecological diseases. Findings of ultrasonographical examinations are operatordependent, leading to the possibility of missing women with smaller intrauterine fibroids.
Additionally, the sample size in this study were small as women who were able to be clinically examined were only from three villages. Further research involving larger sample sizes are necessary to validate the findings in this study.

Conflict of interest
All authors declare no conflict of interest.

Authors' contributions
Author Bing-Jie Wu designed the study, performed the statistical analysis, wrote the