This case-control study recruited women with and without endometriosis aged 20-50 years in 2017 from April to December. The participants in the case group were selected from among women with endometriosis visiting a hospital over the past two years who had undergone laparoscopy and open surgery with a histologic diagnosis of endometriosis and information from medical files in the hospital. Participants in the control group were selected from women of reproductive age (20-50 years) visiting the same centre for other problems, e.g., vaginitis or a medical visit, in whom the absence of endometriosis was confirmed by a gynaecologist based on their signs and symptoms, and considering inclusion and exclusion criteria.

Inclusion criteria: Age of 20-50 years; diagnosis of endometriosis by open surgery or laparoscopy and histologic diagnosis of endometriosis or the presence of endometrioma (case group); being Iranian; being married; absence of endometriosis (control group); willingness to participate; no history of infertility (control group); no history of tubectomy (control group).

Exclusion criteria: Being menopausal (amenorrhea for over a year); endometriosis in the surgical site or the involvement of remote areas, e.g., lungs or brain; suspected of endometriosis or endometrioma (control group); having Polycystic Ovarian Syndrome (PCOS); breast, ovarian, or endometrial cancer; the presence of chronic pelvic pain; having any life-threatening disease.

The sample size was calculated based on the results of a pilot study on 150 participants and considering an Odds Ratio to be detected about 1.8, was determined as n₁=185 (case group), n₂=370 (control group), and n=555 (total) (with the case-to-control ratio of 1:2).

This study was confirmed by the Ethics Committee of Tabriz University of Medical Sciences (Ethics Code: 5/D1003687). Afterwards, data collection was started in Alzahra hospital, Tabriz, which is referral gynaecology and midwifery hospital in NorthWest Iran. By reviewing medical records, those with a confirmed histologic diagnosis of endometriosis through laparoscopy or open surgery were identified over the past two years and their addresses and telephone numbers were extracted from their records. They were contacted via telephone and briefed about research objectives and method, inclusion and exclusion criteria were checked, and they were invited to participate in the study. For those who were willing to take part, questionnaires were filled in by the researchers through interviews. After sampling was done in the case group, control group members were selected through purposive sampling from those visiting the gynaecology clinic of the same centre for other problems, e.g., vaginitis or a medical visit, and did not have endometriosis as diagnosed based on symptoms by a gynaecologist colleague. Research objectives and method were explained to them. For those who were willing to participate, inclusion and exclusion criteria were checked, and in case they met the criteria, they were recruited and questionnaires were completed by the researcher through interviews. Informed consent forms were obtained from all participants, and those in case and control groups were matched for age ±2 years.

Data were collected by the researcher through interviews and using researcher-made questionnaires based on previous studies, highlighting socio-demographic and health behaviours characteristics.

The socio-demographic characteristics questionnaire included questions on age; level of education, level of income, occupation, and endometriosis in first-degree relatives (mother, sisters, and aunts); smoking and alcohol use.

The health behaviour questionnaire included questions on the use of cloth menstrual pads, tampons, sanitary napkins, bathing, genital washing, vaginal washing, and swimming with or without tampons during menstruation.

To determine the content and face validity of the questionnaires, they were given to 10 faculty members and corrections were applied based on their opinions.

Data were analysed in SPSS 21.0 software. Socio-demographic and health behaviours characteristics during menstruation were described using descriptive statistics including frequency (percentage). To determine the association between health behaviours during menstruation and endometriosis, chi-squared test was used in the univariate analysis, and conditional logistic regression was employed in the multivariate analysis to control confounding variables (level of education, level of income, occupation, cycle length, cycle interval, pregnancy number). As no woman in the control group reported a history of this disease in her first-degree relatives; so the family history was not included in the multivariate regression as a confounding factor. Sensitivity analysis was also used to control participants who used both cloth menstrual pads and sanitary napkins during menstruation. In this analysis, the odds ratio and confidence interval was set at 95%, and p<0.05 was considered significant.

In this study, 185 women with endometriosis and 370 women without endometriosis were analysed. Mean age of participants was 35.21±7.09 years in the case group and 35.28±7.03 years in the control group. In terms of educational level; the percentage of academic women in the case group was twice as high as in the control group. Moreover, 32 (17.3%) women in the case group and 10 (2.7%) women in the control group were employed. In the case group, 13 (7%) women reported a history of endometriosis in their mothers and sisters, and 7 (3.8%) women reported this in their aunts, while no woman in the control group reported a history of this disease in her first-degree relatives. Only one woman in the control group had a history of smoking, and no one in either group had a history of alcohol use [Table/Fig-1].

Based on the multivariate logistic regression analysis, the odds of developing endometriosis was about two times higher in those who used cloth menstrual pads during menstruation than those who did not {OR (95% CI)=1.97 (1.12 to 3.44); p=0.017}. There was a significant difference between case and control groups in terms of mean days of using cloth menstrual pads based on the multivariate logistic regression analysis. Thirty-four participants in the case group (77.3%) and 25 participants in the control group (40.3%) used cloth menstrual pads for more than five days during menstruation {OR (95% CI)=6.99 (1.16 to 41.93); p=0.033}. There was also a significant difference between the two groups in terms of the average number of cloth pads used per day; 16 participants in the case group (36.4%) and 12 in the control group (19.4%) used more than three cloth menstrual pads per day during menstruation {OR (95% CI)=6.48 (1.30 to 32.23); p=0.022}.

Based on the multivariate logistic regression analysis there was no significant difference between two groups regarding the use of sanitary napkins and the average number of days in which they used sanitary napkins as well as the average number of sanitary napkins used per day after analysing the data; respectively {OR (95% CI)=1.45 (0.56 to 3.76); p=0.436}, {OR (95% CI)=0.59 (0.32 to 1.07); p=0.085}, {OR (95% CI)=0.65 (0.41 to 1.03); p=0.068}.

Because 76 participants (13.7%) in case and control groups used both cloth menstrual pads and sanitary napkins during menstruation, authors performed sensitivity analysis. Like multivariate logistic regression analysis the results of sensitivity analysis showed that the odds of developing endometriosis were about two times higher in those who used cloth menstrual pads during menstruation than those who did not {OR (95% CI)=1.96 (1.07 to 3.61; p=0.029}, and there was no significant difference between case and control groups in terms of using sanitary napkins during menstruation {OR (95% CI)=1.30 (0.49 to 3.48); p=0.591}.

Based on multivariate logistic regression analysis there was a significant difference between case and control groups in terms of bathing during menstruation {OR (95% CI)=1.74 (1.03 to 2.94); p=0.037}. Moreover, there was a significant difference between the two groups in terms of water temperature in the bathroom based on the multivariate logistic regression analysis. The odds of developing endometriosis were approximately five times higher in women who took hot baths during menstruation than those who did not {OR (95% CI)=4.78 (2.22 to 10.30); p>0.001}. Thirty-four participants in the case group (22.2%) and 15 in the control group (5.5%) reported taking hot baths during menstruation. Ten women in the case group (5.4%) and two in the control group (0.5%) performed vaginal washing during menstruation. Based on multivariate logistic regression analysis there was a significant difference between case and control groups in terms of vaginal washing {OR (95% CI)=7.20 (1.32 to 39.22); p=0.022}.

Only one woman in each group reported using tampons in the swimming pools during menstruation. Four women in the case group and one in the control group reported swimming in pools without tampons during menstruation. Also, two participants in the case group reported swimming without tampons often/sometimes; two participants in the case group reported rarely swimming without tampons during menstruation, and one woman in the control group reported rarely swimming without a tampon during menstruation.

All the data mentioned above are shown in [Table/Fig-2,3].

The aetiology and pathogenesis of endometriosis are not well-understood, and the theory of retrograde menstruation has a broader acceptance than other theories. Thus, it is believed that some health behaviours such as using tampons and douches or vaginal washing during menstruation may increase the chance of developing endometriosis. Therefore, it is important to identify the predisposing factors of retrograde menstruation. This study was the first in Iran to investigate the correlation between endometriosis and health behaviours (vaginal washing and swimming with or without tampons) during menstruation as predisposing factors of retrograde menstruation. The results of multivariate regression analysis indicated that the risk of developing endometriosis was about twice higher in women who used cloth menstrual pads during menstruation than those who did not. Moreover, women with endometriosis used cloth menstrual pads for more than five days and more than three times a day.

Furthermore, the risk of endometriosis was approximately two times higher in those women who stated they took bath during menstruation and five times higher in women who took a hot bath during menstruation compared to those who state they did not.

The chance of developing endometriosis was approximately seven times higher in those women who had vaginal washing during menstruation compared to those who state they did not.

The reason for these results is unclear because no studies have been conducted on this issue so far. A possible answer may be that the absorption capacity of cloth menstrual pads is lower than sanitary napkins with absorbent materials, increasing the probability of retrograde menstruation due to blood stasis in the vagina.

Although there are studies that demonstrate the effect of hot baths on decreasing dysmenorrhea and dyspareunia associated with endometriosis [2,4], there is no study that shows the effect of hot baths as a predisposing factor of endometriosis during menstruation. It seems that taking hot baths and the presence of steam may be predisposing factors, increasing the chance of retrograde menstruation during menstruation. Studies have proved the effect of hot baths on muscle relaxation [7]. Therefore, the relaxation of the wall of uterine vessels might lead to an increase in menstrual bleeding, which is a major risk factor for endometriosis and increases the flow of retrograde menstruation [6,11]. Iranian traditional medicine does not support bathing during menstruation, and it is believed to be associated with many complications for women, especially in those days when the volume of bleeding is high. All these subjects may increase retrograde menstruation but their mechanisms are unclear and further studies are required.

So far, few studies have focused on the correlation between some health behaviours and endometriosis. A case-control study conducted by Meaddough EL et al., in the United States investigated the effect of sexual activity, orgasm, and health behaviours during menstruation on endometriosis [6]. The findings showed the majority of participants in the two groups never performed vaginal douching. This finding is not consistent with the findings of the present study. In the present study, case and control groups were significantly different in terms of performing vaginal douching or vaginal washing during menstruation based on multivariate logistic regression analysis, but the confidence interval was very wide, probably due to the low number of participants reporting vaginal washing during menstruation so that definitive conclusion is not possible. This study showed a negative correlation between the use of tampons during menstruation and endometriosis and a positive correlation between the use of sanitary napkins and endometriosis. Meaddough EL et al., concluded that tampons may be more absorbent than sanitary napkins [6]. This is of course not proven yet since few studies have been conducted on the presence of toxic substances such as dioxin in sanitary napkins and tampons or the toxicity of dioxin and its correlation with endometriosis [9,12,13]. In a case-control study conducted by Heilier JF et al., in Belgium, the results indicated no significant difference between the two groups regarding the use of tampons and sanitary napkins [14]. A study by Darrow SL et al., showed that using tampons for over 14 years increases the risk of developing endometriosis by four times [15]. However, several studies reported no statistically significant difference between case and control groups regarding the use of tampons [14,16]. In the present study, there was no significant difference between case and control groups in the use of sanitary napkins based on the multivariate regression analysis. Also, few participants reported the use of tampons, which is perhaps due to the cultural differences as Iranian women are not familiar with a tampon and its usage.

In this study, an attempt was made to select women with and without endometriosis based on specialised and precise diagnosis. Furthermore, factors were examined while controlling confounding variables.

In the present study, the case groups were selected from women with confirmed diagnosis of endometriosis through laparoscopy or laparotomy, but the control group were selected from women in whom the absence of endometriosis was confirmed by a gynaecologist based on their signs and symptoms, and considering inclusion and exclusion criteria. The criterion for the definite diagnosis of endometriosis is histologic diagnosis through laparoscopy or laparotomy therefore future studies may choose to select both case and control participants from women in whom the presence or absence of endometriosis is confirmed by laparoscopy or laparotomy. Also, in the present study to determine the content and face validity of the questionnaires, they were given to 10 faculty members and corrections were applied based on their opinions. Authors recommend performing further studies with more validated questionnaire on this topic.

In this study validity of the questionnaire was only confirmed through face and content validity qualitatively and the quantitative indices such as Content Validity Index (CVI) and Content Validity Ratio (CVR) weren’t calculated. Also, considering the criterion for the definite diagnosis of endometriosis is histologic diagnosis through laparoscopy or laparotomy, one of the other limitations of this study is that the control group wasn’t selected based on histologic diagnosis. Thus, future studies should be done selecting both case and control participants from women in whom the presence or absence of endometriosis is confirmed by laparoscopy or laparotomy.

Based on the results of this study and owing to an increased prevalence of endometriosis compared to the past as well as considering the numerous complications of endometriosis, e.g., infertility, it seems logical to recommend that healthcare teams and those in charge in healthcare centres educate women of reproductive age about the observance of health behaviours during menstruation. Nevertheless, further studies are required.

Funding: This study was funded by the Research Deputy of Tabriz University of Medical Sciences.