2022 Volume 69 Issue 12 Pages 1437-1445
Dienogest (DNG) is widely used to treat dysmenorrhea associated with estrogen-dependent diseases such as endometriosis and adenomyosis. DNG becomes unnecessary after menopause when estrogen secretion declines drastically. However, there are no clear criteria for when to halt DNG in perimenopausal patients. Menstruation and dysmenorrhea often resume after discontinuation due to approaching menopause. This case-control study used serum estradiol and follicle-stimulating hormone (FSH) levels to predict whether menstruation would resume in perimenopausal women after discontinuation of DNG. The study enrolled patients aged ≥40 years with endometriosis and/or adenomyosis and who had either completed oral DNG therapy (DNG group) or had spontaneous menopause without hormone therapy (control group). We assessed estradiol and FSH values before DNG termination or the final menstrual period. DNG group members that resumed menstruation after DNG termination (D (+) group, n = 17) had significantly higher estradiol and lower FSH levels than those who did not (D (–) group, n = 22) up to four months before DNG termination but not from four to 12 months. Estradiol and FSH levels were not significantly different between the D (–) and control groups. Receiver operating characteristic curves created from the estradiol and FSH values indicated that menstruation resumed when levels were ≥17 pg/mL and <100 mIU/mL, respectively. In contrast, menstruation did not resume in cases of estradiol ≤20 pg/mL and FSH >80 mIU/mL. The study results provide useful criteria for deciding when to terminate DNG in perimenopausal patients that consider their tolerance for resuming menstruation. Applications to menopause-inducing therapy for uterine fibroids and other conditions are anticipated. Further large-scale studies are needed.
ACCORDING TO a recent meta-analysis, the mean age of menarche is 12.9 years and that of natural menopause is 50.5 years. The age of menarche has been declining, while the age of natural menopause has not changed significantly [1]. However, it is difficult to accurately predict the occurrence of natural menopause. Studies have attempted to estimate the timing of menopause based on changes in the estradiol and follicle-stimulating hormone (FSH) levels [2-6]. However, these predictions were not very accurate, and although anti-Müllerian hormone (AMH) levels have recently been used for prediction, there is still no established method for predicting the timing of menopause [7, 8].
Endometriosis is a chronic, inflammatory, estrogen-dependent disease characterized by endometrium-like tissue found outside the uterus [9]. It causes dysmenorrhea, chronic pelvic pain, and lowers the quality of life (QOL) [10]. Adenomyosis is a disease characterized by endometrium-like tissue found in the myometrium, which lowers the QOL due to menorrhagia and dysmenorrhea. Its treatment is based on its similarities with endometriosis and estrogen dependence [11].
Dienogest (DNG) is a fourth-generation progestin used to treat dysmenorrhea associated with endometriosis and adenomyosis because of its ability to inhibit ovarian function and suppress endometrial hyperplasia [12]. It has been shown to be effective for pain associated with endometriosis and adenomyosis [13, 14]. Lower abdominal pain has been shown to adversely affect QOL, and DNG relieves pain associated with endometriosis and adenomyosis with no adverse effects, even with long-term administration or large doses [13, 15-18]. The European Society of Human Reproduction and Embryology guidelines recommend DNG for endometriosis pain but do not address areas that are important in clinical practice, such as how long it should be used and when to end therapy [12]. However, the pain-suppressing effects of this drug can be expected to improve QOL and increase productivity, which has justified its use in endometriosis, as long-term management is recommended for this condition [19]. In clinical practice, patients often request this. While oral DNG mildly suppresses ovarian function, there have been no attempts to estimate the timing of menopause or to examine changes in hormone dynamics in perimenopausal patients who are taking DNG continuously. Based on the above, it is reasonable for perimenopausal women aged ≥45 years to continue taking DNG even though the progression of menstruation-dependent diseases, such as endometriosis, has largely ended. However, it is difficult to identify the criteria for continued administration. Gonadotropin-releasing hormone (GnRH) agonists can be used to induce menopause in women with uterine fibroids. Therefore, if the timing of menopause can be estimated in patients taking DNG, they could continue taking it until menopause approaches. Halting administration after that point would maximize the therapeutic effect of DNG on dysmenorrhea associated with endometriosis and adenomyosis [20, 21].
Existing data on hormonal changes in women taking DNG are from relatively young, menstruating women. Although predicting the time of menopause from menstrual patterns is difficult in all menopausal women, women receiving DNG may face complications as DNG has a mild inhibitory effect on ovarian function. However, for the aforementioned reasons, predicting the timing of menopause from the hormone levels in women taking DNG would be useful to determine whether to continue or stop treatment. Therefore, the objective of this study was to predict whether menstruation would resume after stopping DNG based on the serum estradiol and FSH levels in the perimenopausal period of women who were taking DNG for endometriosis or adenomyosis.
This case-control study used data from two previous studies conducted at the University of Tokyo Hospital. The first was a retrospective study of 57 patients with endometriosis or adenomyosis. This cohort included patients examined at our hospital’s outpatient department of gynecologic surgery from January 2011 to December 2016. The second was a prospective cohort study of 207 women aged ≥40 years with a history of endometriosis or endometrial lesions at the study initiation. This study was conducted from December 2015 to January 2019. Both studies were approved by the research ethics committee of the University of Tokyo, and all participants provided written consent (approval no. 0324). The participants from these two cohorts who met the following criteria were included in this study and classified into two groups. The criteria for inclusion were age ≥40 years and the presence of dysmenorrhea associated with endometriosis or adenomyosis, which was treated with oral DNG (2 mg/day) for ≥12 months. The following exclusion criteria were used: (1) endocrinological diseases (i.e., diabetes and chronic kidney disease), smokers, and steroid administration; (2) history of bilateral salpingo-oophorectomy or total hysterectomy; and (3) absence of a blood test performed at <4 months and 4–12 months before DNG termination or the final menstrual period (FMP). The control group (C group) comprised patients who were not receiving DNG or other hormone therapies and had blood hormone levels measured at <4 months before their FMP. The DNG group comprised patients aged ≥40 years who had been receiving DNG for at least 1 year and had blood hormone levels measured at <4 months before DNG termination. The DNG group was further subdivided into two groups based on whether menstruation resumed after stopping DNG. The D (+) group comprised patients who had at least one menstrual period after stopping DNG. The D (–) group comprised patients who did not menstruate and immediately went into menopause after stopping DNG. In total, 47 patients were included in the study. In the D (+) and D (–) groups, we collected data on the age at DNG termination, the interval between DNG termination and the blood test, duration of DNG administration, history of ovarian surgery, and estradiol (chemiluminescence enzyme immunoassay [CLEIA] method) and FSH (CLEIA method) values from <4 months and 4–12 months before DNG termination. In the C group, we collected data on the age at FMP, the interval between the FMP and the blood test, history of ovarian surgery, and estradiol and FSH values from <4 months and 4–12 months before the FMP. Cases with values below the minimum level of estradiol detection (<12 pg/mL) were recorded as 12 pg/mL. The FSH/estradiol ratio was calculated for use in the analysis. For the blood sampling interval, values of 0 were at DNG termination, with negative values before termination.
Statistical analysesThe data were analyzed for normality using the Shapiro–Wilk and Kolmogorov–Smirnov tests. Based on these results, normally distributed data were expressed as mean ± standard deviations, and non-normally distributed data were expressed as medians (ranges). Two-group comparisons were performed using the t-test and Mann–Whitney U test for normal and non-normal data, respectively. Fisher’s exact test was used for categorical data. Three-group comparisons were performed using analysis of variance and the Kruskal–Wallis test for normal and non-normal data, respectively, and the Tukey or Dunn methods were used for post-hoc analysis.
Based on the results of the two-group comparisons of whether menstruation resumed after DNG termination, logistic regression analysis was performed using estradiol and FSH values to estimate whether menstruation would resume after DNG termination. To calculate a more practical score for predicting the resumption of menstruation after DNG termination, receiver operating characteristic (ROC) curves were created from the estradiol and FSH values to examine the diagnostic accuracy. In addition, different cut-off values were used to calculate sensitivity, specificity, and diagnostic accuracy. All statistical analyses were performed using the GraphPad Prism software (GraphPad Software, Inc., CA) version 8 for Windows. Statistical significance was set at p < 0.05.
Of the patients with dysmenorrhea associated with endometriosis and adenomyosis, 17 menstruated after DNG termination (D (+) group) and 22 did not (D (–) group). Eight patients with a history of endometriosis did not take DNG (C group). The characteristics of the D (+), D (–), and C groups are shown in Table 1. There were no significant differences in the age at FMP or DNG termination or in the interval between the FMP or DNG termination and the hormone test. No significant differences were observed in the history of ovarian surgery. There was no significant difference in the duration of DNG use between the D (+) and D (–) groups.
| D (+) group (N = 17) | D (–) group (N = 22) | Control (N = 8) | p | |
|---|---|---|---|---|
| Age at DNG termination/menopause (years) | 49 (42–55) | 49 (44–52) | 46.5 (44–51) | 0.3490 |
| Interval between DNG termination/ FMP and hormone test (months) |
–2 (–3 –0) | 0 (–3 –0) | 0 (–3 –2) | 0.1763 |
| Duration of DNG intake (months) | 73.1 ± 33.8 | 78.5 ± 27.2 | — | — |
| Past history of ovarian surgery | 11 (65%) | 17 (77%) | 7 (88%) | 0.4562 |
The control group was defined as the menopausal group of women with endometriosis who did not receive DNG.
Data were presented as median (range), mean ± SD or number of cases (percentage).
Negative values indicate before the end of DNG in the interval between DNG termination/FMP and hormone tests.
DNG: Dienogest, FMP: final menstrual period, E2: estradiol, FSH: follicle-stimulating hormone
The comparison of estradiol and FSH levels at <4 months and 4–12 months before DNG termination between the D (+) and D (–) groups is shown in Fig. 1. The estradiol and FSH levels at 4–12 months before DNG termination were not significantly different between the D (+) and D (–) groups (p = 0.16, p = 0.16, respectively). The estradiol levels at <4 months before DNG termination were 27.1 (17.0–202.0) pg/mL and 15.6 (12.0–26.5) pg/mL in the D (+) and D (–) groups, respectively. The estradiol levels were significantly higher in the D (+) group than in the D (–) group (p < 0.001). The FSH levels at <4 months were 50.4 (8.8–96.7) mIU/mL and 76.8 (28.3–135.5) mIU/mL in the D (+) and D (–) groups, respectively. The levels in the D (+) group were significantly lower than those in the D (–) group, and the FSH/estradiol ratio was significantly lower in the D (+) group compared with the D (–) group (p = 0.004 and p < 0.001, respectively; data not shown). The comparison of hormone levels between the D (–) and C groups is shown in Fig. 2. There were no significant differences in the estradiol levels, FSH levels, or FSH/estradiol ratio between the D (–) and C groups.
Hormone levels at different times between D (+) and D (–) groups before DNG termination
Comparison of hormone levels between the D (–) and D (+) groups at 4–12 months before DNG termination and at <4 months before DNG termination. Each graph shows medians and interquartile ranges. (A) E2 value of the D (–) group and D (+) group at 4–12 months before DNG termination. There was no significant difference between the two groups. (B) E2 value of the D (–) and D (+) groups at <4 months before DNG termination. The E2 value in the D (–) group was significantly lower than that in the D (+) group. (C) FSH value of the D (–) and D (+) groups at 4–12 months before DNG termination. There was no significant difference between the two groups. (D) FSH value of the D (–) and D (+) groups at <4 months before DNG termination. The FSH value in the D (–) group was significantly higher than that in the D (+) group.
**: p < 0.01
***: p < 0.001
D (+), the group that resumed menstruation after DNG termination; D (–), the group that did not resume menstruation after DNG termination
DNG, dienogest; E2, estradiol; FSH, follicle-stimulating hormone
Hormone levels in the D (–) and control groups at <4 months before DNG termination.
Each graph shows medians and interquartile ranges. (A) E2 value of the D (–) and control groups at <4 months before DNG termination. There was no significant difference between the two groups. (B) FSH value of the D (–) and control groups at <4 months before DNG termination. There was no significant difference between the two groups. (C) FSH/E2 ratio of the D (–) and control groups at <4 months before DNG termination. There was no significant difference between the two groups.
D (–), group that did not resume menstruation after DNG termination
DNG, dienogest; E2, estradiol; FSH, follicle-stimulating hormone
The results of logistic regression analysis to estimate whether menstruation would resume after DNG termination were as follows: estradiol (odds ratio [OR], 1.44; 95% confidence interval [95% CI], 1.16–2.03; p = 0.007), FSH (OR, 0.96; 95% CI, 0.93–0.99; p = 0.01), and FSH/estradiol ratio (OR, 0.47; 95% CI, 0.25–0.71; p = 0.003). The areas under the ROC curves (AUC) were 0.91, 0.76 and 0.86 for estradiol levels, FSH levels, and FSH/estradiol ratio, respectively.
To calculate a more practical score for predicting whether menstruation would resume after DNG termination, ROC curves were created from the estradiol and FSH values. The ROC curves for estradiol, FSH, and the FSH/estradiol ratio are presented in Fig. 3. The AUC of estradiol was the largest at 0.87. The sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, and negative likelihood ratio for different cut-off values of estradiol and FSH levels are shown in Table 2. From the ROC curves of estradiol and FSH, menstruation resumed after DNG termination at estradiol ≥20 pg/mL and FSH <56 mIU/mL. Estradiol and FSH values were combined to calculate a more practical cut-off value with better diagnostic accuracy. Cut-off values for certainty that menstruation would resume were estradiol ≥17 pg/mL and FSH <100 mIU/mL. Cut-off values for certainty that menstruation would not resume were estradiol <20 pg/mL and FSH ≥80 mIU/mL.
ROC curves of E2 levels, FSH levels, and FSH/E2 ratio
ROC, receiving operating characteristic; E2, estradiol; FSH, follicle-stimulating hormone
| Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | LR+ | LR– | |
|---|---|---|---|---|---|---|
| E2 ≥20 | 70.6 | 95.5 | 92.3 | 80.8 | 15.5 | 0.3 |
| E2 ≥17 | 100.0 | 27.3 | 51.5 | 100.0 | 1.4 | 0.0 |
| FSH <40 | 41.2 | 95.5 | 87.5 | 67.7 | 9.1 | 0.6 |
| FSH <56 | 58.8 | 68.1 | 58.8 | 68.1 | 1.8 | 0.6 |
| FSH <100 | 100.0 | 22.7 | 50.0 | 100.0 | 1.3 | 0.0 |
| E2 ≥20 and FSH <80 | 64.7 | 100.0 | 100.0 | 78.6 | - | 0.4 |
| E2 ≥18 and FSH <90 | 88.2 | 72.7 | 71.4 | 88.9 | 3.2 | 0.2 |
| E2 ≥17 and FSH <100 | 100.0 | 68.2 | 70.8 | 100.0 | 3.1 | 0.0 |
E2, oestradiol; FSH, follicle-stimulating hormone; NPV, negative predictive value; PPV, positive predictive value; LR+, positive likelihood ratio; LR–, negative likelihood ratio.
In order to see the changes in serum hormone levels during long-term treatment with DNG, Fig. 4 shows the measured hormone levels during DNG treatment in the D (–) and D (+) groups, categorized into 4 months, 4–12 months, 20 months, and every 20 months thereafter, respectively, before DNG termination. Fig. 4A–D show the changes in estradiol in the D (+) group, FSH in the D (+) group, estradiol in the D (–) group, and FSH in the D (–) group, respectively. No differences in the hormone levels between the D (–) and D (+) groups were evident. In Fig. 5, the estradiol and FSH values of the D (–) and D (+) groups during DNG treatment were categorized and compared in the same way as in Fig. 4. In Fig. 5A, no differences in estradiol levels between the D (–) and D (+) groups were evident. In contrast, Fig. 5B showed that mean FSH levels were consistently higher in the D (–) group than in the D (+) group, although both groups showed an increasing trend, with a sharp increase in the 12 months before DNG termination.
Time-course change of hormone levels during the treatment in each patient
Each graph shows the hormone measurements for an individual patient at each time point.
(A) (B) E2 and FSH levels at 4 months, 4–12 months, 20 months, and every 20 months thereafter before DNG termination for each patient in the D (+) group. (C) (D) E2 and FSH levels at 4 months, 4–12 months, 20 months, and every 20 months thereafter before DNG termination for each patient in the D (–) group.
D (+), the group that resumed menstruation after DNG termination; D (–), the group that did not resume menstruation after DNG termination
DNG, dienogest; E2, estradiol; FSH, follicle-stimulating hormone
Trends in hormone levels in D (+) and D (–) groups during DNG treatment
Each graph shows the mean and standard deviation of hormone values at each time point.
(A) E2 values at each time point for patients in the D (+) and D (–) groups, at 4 months, 4–12 months, 20 months, and every 20 months thereafter before DNG termination. (B) FSH values at each time point for patients in the D (+) and D (–) groups, at 4 months, 4–12 months, 20 months, and every 20 months thereafter before DNG termination.
DNG, dienogest; E2, estradiol; FSH, follicle-stimulating hormone
The objective of this case-control study was to predict whether menstruation would resume in a group of patients who continued taking DNG during the perimenopausal period for indications such as endometriosis and adenomyosis. The FSH and estradiol levels were used to identify hormone dynamics to aid decision-making on when DNG should be terminated. Recently, AMH has been used to predict the timing of natural menopause. AMH is produced from preantral follicles and reflects the ovarian reserve. Conventional methods cannot be used to accurately predict the age of menopause because measurements become less accurate in the late reproductive stage, as AMH declines with age and cannot be detected after menopause. In contrast [8, 22], FSH and estradiol levels, which are widely used to predict the timing of menopause, are problematic because they fluctuate in the perimenopausal period, although they are relatively popular and easy to measure [23]. Our results showed that the estradiol levels were significantly lower, and the FSH levels were significantly higher in the group that did not resume menstruation after DNG termination than in the group that resumed menstruation. Further, this change was observed <4 months before DNG termination but not at 4–12 months before termination. The cut-off values for resumption of menstruation after DNG termination calculated from the ROC curves were ≥20 pg/mL for estradiol and <56 mIU/mL for FSH.
The study’s finding that estradiol levels were significantly lower and FSH levels were significantly higher in the group that did not resume menstruation after DNG termination is reasonable as it considers the hormonal changes that occur during natural menopause [4]. The results of previous research predicting the timing of menopause largely support our results, as the geometric mean (95% CI) of the FMP was 48.4 (39.0–60.1) IU/L for FSH and 30.8 (22.6–42.2) pg/mL for estradiol [24].
DNG does not suppress the basic secretion of gonadotropins, such as FSH and luteinizing hormone (LH), and has a moderate effect on the suppression of estrogen secretion [16, 25, 26]. A study of patients in Japan who were taking DNG (2 mg/day) recorded estradiol levels of 42.1 ± 50.7 (24.0) pg/mL (mean ± SD [median]) after 16 weeks [13]. In this study, the estradiol value at <4 months before DNG termination in the group that resumed menstruation was similar to that of the women taking DNG in the previous study. The estradiol value in the group that did not resume menstruation in the present study tended to be lower than that in the previous study. In contrast, the FSH value in the group that did not resume menstruation was at the same level as the FMP of the natural menopausal women in the previous study. There was no significant difference between the FSH value in the group that did not resume menstruation and that in the control group. These results are considered reasonable considering that DNG does not suppress FSH secretion but suppresses endogenous estradiol secretion. Thus, it was speculated that cessation of menstruation happened at similar times in both D (–) and control groups, and the endocrinological background of these groups did not significantly differ. In the present study, some patients in the group who did not resume menstruation had estradiol values below the minimum limit of measurement. As these cases were assigned the minimum limit of measurement as the estradiol level, we presume that the actual estradiol level of this group was even lower. In Fig. 5, which compares the D (–) and D (+) groups, the pattern of change in estradiol levels in the D (–) and D (+) groups was similar, but FSH levels were consistently higher in the D (–) group than in the D (+) group, and a sharp increase was observed in the 12 months before DNG termination. It is reasonable to assume that the mild estradiol suppression effect of DNG modifies the changes in estradiol levels, resulting in no clear difference between the two groups. In contrast, the rapid increase in FSH levels in the D (–) group compared to the D (+) group in the 12 months before DNG termination suggest that DNG does not suppress gonadotropins.
Previous studies did not follow many patients on long-term DNG up to the perimenopausal period. Thus, there has been little analysis of the endocrine dynamics of patients who continue receiving DNG until menopause. Therefore, the timing of menopause and the changes in hormone levels in the perimenopausal period were not known in such patients. In a report of two patients who stopped taking DNG when estradiol and FSH levels reached the menopausal level (estradiol ≤39.5 pg/mL, FSH 25.8–134.8 mIU/mL), menstruation did not resume in either patient. The present study is significant as it analyzed and compared changes in the estradiol and FSH levels in patients taking DNG during the perimenopausal period based on whether menstruation resumed [27].
In the present study, the cut-off values at <4 months before DNG termination in the group that did not resume menstruation were estradiol ≥20 pg/mL and FSH <56 mIU/mL. According to a previous report, the cut-off value for FSH during menopausal transition and menopause is 45 mIU/mL [2]. Furthermore, although FSH >40 IU/L is used as an indicator for post-menopause, it has been reported that the bleeding pattern is more useful for predicting the time until the FMP than FSH levels [28]. As the menstrual bleeding pattern cannot be used as a predictor of menopause in patients taking DNG, the findings of the present study could be useful in predicting menopause in patients taking DNG. Furthermore, reduced bone mineral density has been indicated as an adverse event of long-term DNG administration; thus, predicting the timing of menopause could be used to avoid long-term administration and, thus, suppress adverse events [16, 29]. In addition, the degree to which a patient can tolerate the resumption of menstruation differs according to the severity of dysmenorrhea symptoms. If symptoms are intolerable, a model with a high specificity level can be used, and the cut-off value from the present study would be useful.
GnRH agonists may be used to induce menopause in patients with hypermenorrhea or dysmenorrhea caused by uterine fibroids [21]. A report on this therapy that used age ≥45 years as well as FSH ≥25 mIU/mL and LH ≥25 mIU/mL on days 3–5 of the menstrual cycle as the criteria found that menstruation did not resume after 24 weeks of treatment of uterine fibroids with leuprorelin acetate in 19 out of 21 women (90%). It was concluded that FSH >25 mIU/mL was needed to successfully induce natural menopause [20, 21, 30]. DNG inhibits estrogen secretion more mildly than GnRH agonists; thus, it is reasonable to obtain a higher FSH cut-off value.
The strengths of this study were the comparison of the estradiol and FSH levels in patients taking DNG during the perimenopausal period and the estimation of cut-off values for resuming menstruation after terminating administration. However, this study had some limitations. First, the sample size was not sufficiently large, especially in the control group. This study did not observe a significant difference in the estradiol levels between the group that did not resume menstruation after DNG termination and the natural menopause group. The inhibitory effect of DNG on estrogen secretion may explain the lower estradiol levels in the group that did not resume menstruation after DNG termination compared to the natural menopause group.
Second, we only analyzed the FSH and estradiol hormone levels. Estradiol and FSH levels exhibit large fluctuations during the perimenopausal period; thus, sampling at one time point may have low reproducibility. However, a highly sensitive AMH test with a lower detection limit has been developed. It has been reported that this test is more useful than FSH levels for predicting the FMP [8, 22]. In the future, we would like to examine more cases in the control group and include AMH to predict the timing of menopause in patients taking DNG.
In this study, we compared a group of perimenopausal patients who resumed menstruation after DNG termination and a group that did not resume menstruation to predict whether menstruation would resume by evaluating the FSH and estradiol values. The estradiol levels were significantly lower, and the FSH levels were significantly higher in the group that did not resume menstruation after DNG termination than in the group that resumed menstruation. Menstruation resumed after DNG termination at estradiol levels ≥20 pg/mL and FSH levels <56 mIU/mL. Cut-off values for certainty that menstruation would resume were estradiol ≥17 pg/mL and FSH <100 mIU/mL. Cut-off values for certainty that menstruation would not resume were estradiol <20 pg/mL and FSH ≥80 mIU/mL. The results of this study may help determine when to terminate DNG in perimenopausal patients in a way that considers patient tolerance of menstruation due to dysmenorrhea or other factors. Large-scale studies should be conducted in the future.
FSH, follicle-stimulating hormone; AMH, anti-Müllerian hormone; QOL, quality of life; FMP, final menstrual period; CLEIA, chemiluminescence enzyme immunoassay; ROC, receiver operating characteristic; OR, odds ratio; LH, luteinizing hormone
We would like to thank Editage (www.editage.com) for English language editing.
This study was supported by the Japan Agency for Medical Research and Development (grant numbers 15gk0210007h0101 and 20gk0210018h0003) and the Ministry of Health, Labour and Welfare (19FB0101). The funders did not play any role in the design of the study; data collection, analysis, and interpretation and writing the manuscript.
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
All procedures performed in the study, involving human participants, were in accordance with the ethical standards of the institutional research committee and the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Ethical approval for this study was obtained from the University of Tokyo Hospital Institutional Review Board (approval number: 0324). Informed consent was obtained from all participants included in the study.
None of the authors have any potential confilcts of interest associated with this research.
