Complementary Chinese Herbal Medicine Treatment is Associated with a Reduction of Surgical Rate in Patients with Dysfunctional Uterine Bleeding: A Propensity-Score Matched Cohort Study

Dysfunctional uterine bleeding (DUB) is defined as abnormal bleeding (in terms of the frequency, duration, and amount of bleeding) without any clear structural or systemic etiology identified. It is a common problem in women of reproductive age. Approximately 30% of women encounter the problem of abnormal uterine bleeding annually, and 25% of gynecologic surgeries are performed to treat this condition. 1 Furthermore, approximately 30–40% of hysterectomies are performed for the treatment of severe dysfunctional bleeding. 2 Uterine bleeding has a substantial impact on women’s quality of life and may have a negative impact on healthcare costs as well as an associated loss of productivity. 3 Overall, it has significant effects on the medical, socioeconomic, and psychological well-being of women. Medical treatments such as NSAIDs, oral contraceptive pills, progestin, danazol, GnRH agonists, and anti-fibrinolytic drugs are used to reduce menstrual flow. 4 , 5 Symptom recurrence after discontinuing the treatment is considered the main disadvantage of the medical therapies. The associated side effects and the necessity for long-term medical treatment may lead to patients’ noncompliance during prolonged medical therapy, which is a factor influencing the decision of surgical management. If these pharmacological options are unsuccessful, surgical procedures of hysterectomy or the levonorgestrel intrauterine system (LNGIUS) (Mirena) are used. 6 Hysterectomy causes complete cessation of menstruation, leading to complete control of the bleeding symptom. However, it requires considerable recovery time and is an expensive, invasive, and irreversible treatment procedure. 7 In addition, hysterectomy may involve post-operative complications including hemorrhage, injury to adjacent organs such as the intestines and bladder, febrile morbidity, life-threatening events, urinary retention, urinary tract infections (UTIs), and wound infections. 8 The most frequent long-term problems reported after surgery include fistula, chronic pain, bladder or bowel dysfunction, early menopause, sexual dysfunction, poor appetite, constipation, back pain, urinary problems, and psychiatric disturbance. 9 Admission to hospital for hysterectomy not only impacts the daily activities of the patients but also extends to patients’ families and their employers. 10 Because surgery may result in long-term complications, negative quality of life, and significant social or economic costs, some women affected by DUB may seek complementary therapies such as traditional Chinese medicine (TCM) consultations in Taiwan. 11 In Taiwan, both conventional Western medicine and TCM are quite popular and considered as mainstream therapies for gynecological disorders. 11–15 The National Health Insurance program reimburses both treatments. 16 From the perspective of Chinese medicine theory, DUB is described as “Ben Lou” in TCM literature. It is regarded as “flooding and spotting” as a disruption of the harmony of Qi and Blood, which could be caused by “Blood Heat”, “Blood Stasis” and “Organ Dysfunction (Spleen, Liver and Kidney)”, and eventually leads to “Blood/Yin Deficiency”. 11 TCM views DUB as a sign of the body’s internal imbalance and the treatments aim to harness the balance of “Yin and Yang” in the body through a combination of herbal medicine, acupuncture, dietary changes, and lifestyle modifications. While this theory fits some of the pathomechanisms of DUB, it is also important to point out that gynecological surgeries also play an important role in clinical management. We therefore are curious about whether there is a space to integrate TCM treatment in the management of DUB. In the meantime, the advantage of complementary TCM among patients of DUB in real-world clinical settings needs some substantial evidence. We previously illustrated the prescription patterns and core prescriptions of Chinese herbal medicine (CHM) for patients with DUB. 11 Yi-Mu-Cao (Leonuri Herba) and Jia-Wei-Xiao-Yao-San were the most commonly prescribed single herb and herbal formula for patients with DUB in Taiwan. In this study, we aimed to investigate the relationship between surgery rate and complementary CHM treatment in patients with DUB. This could help determine the effects of TCM for patients with DUB and provide valuable information for clinical gynecologists.

The National Health Insurance (NHI) program was established in Taiwan in 1995. The program is highly representative of Taiwan’s general population because the reimbursement policy is universal and mandatory. It covered more than 99% of Taiwanese residents. The data source of our study was the Longitudinal Health Insurance Database 2000 (LHID 2000), a real-world data derived from the National Health Insurance Research Database (NHIRD). 17 The LHID 2000 contains all the original claims data of 1 million beneficiaries randomly sampled from the registry of all beneficiaries in 2000. 17 We conducted a nationwide, population-based, 1:1 propensity score-matched cohort study by analyzing data derived from the LHID 2000. The sampled patients exhibited no significant differences in age, birth year, or average insured payroll-related costs compared to the general population. We acquired the diagnostic codes in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) format. This study was approved by the Research Ethics Committee of China Medical University and Hospital, Taiwan (CMUH104-REC2-115). Patients newly diagnosed with dysfunctional uterine bleeding (ICD-9-CM code 626.8) (n=51,354) with at least 2 claims from January 1, 1997, to December 31, 2010, were selected from the database ( Figure 1 ). The exclusion criteria included age younger than 18 years, incomplete information on age and sex, and withdrawal from the NHIRD during the follow-up period. Patients who received hysterectomy or endometrial ablation before the diagnosis of DUB were also excluded. To exclude the non-DUB reasons for hysterectomy, patients with a diagnosis of cervix uteri cancer (ICD-9-CM: 180.9, 233.1; A code: A120, A16), endometrial cancer (ICD-9-CM: 182), ovarian cancer (ICD-9-CM: 183, A code: A123), leiomyoma (ICD-9-CM: 218, A code: A152), and endometriosis (ICD-9-CM: 617) within 3 months before or after the diagnosis of DUB were also excluded. Finally, 43,027 patients with newly diagnosed DUB were included. Figure 1 Flow recruitment chart. We identified the newly diagnosed dysfunctional uterine bleeding (DUB) patients from 1997 to 2000 from one million randomly selected subjects of the Longitudinal Health Insurance Database 2000 (LHID 2000) in Taiwan. After excluding patients according to the excluding criteria, we included 43,027 patients and separated them into CHM users (n=38,324) and non-CHM users (n=4,703) according to whether they received TCM treatment or not after the initial diagnosis of DUB. After performing 1:1 propensity score matching, there were 4642 patients in the CHM and non-CHM cohorts. Flow recruitment chart. We identified the newly diagnosed dysfunctional uterine bleeding (DUB) patients from 1997 to 2000 from one million randomly selected subjects of the Longitudinal Health Insurance Database 2000 (LHID 2000) in Taiwan. After excluding patients according to the excluding criteria, we included 43,027 patients and separated them into CHM users (n=38,324) and non-CHM users (n=4,703) according to whether they received TCM treatment or not after the initial diagnosis of DUB. After performing 1:1 propensity score matching, there were 4642 patients in the CHM and non-CHM cohorts. Patients who received CHM from their initial diagnosis of DUB to December 31, 2010, were identified as the CHM users (n=38,324). Those who never received CHM during the same period were identified as non-CHM users (n=4703). Propensity score matching was used to select a comparable CHM group and non-CHM group. We performed a 1:1 propensity-score match based on each subject’s age (per 5 years), comorbidities, drugs used, childbirth status, duration from the diagnosis year of DUB and index year through multiple logistic regression analysis to reduce the selection bias. Ultimately, equal numbers (n=4642) of patients in the CHM group and non-CHM group were analyzed and followed up until December 31, 2013. Baseline comorbidities were considered present if ICD-9-CM codes appeared two or more times in outpatient or inpatient claims before the initial diagnosis of DUB, which included dysmenorrhea (ICD-9-CM: 625.3), inflammatory disease of ovary (ICD-9- CM: 614), anemia (ICD-9- CM: 280–285, A200), menorrhagia (ICD-9- CM: 626.2), female infertility (ICD-9-CM: 628, A376), and hypotension (ICD-9-CM: 458.0, 458.1, 458.9). The drugs used included progesterone, estrogen, danazol, tranexamic acid, nonsteroidal anti-inflammatory drug (NSAID) and gonadotropin releasing hormone (GnRH) agonists. The date of first CHM treatment after a new diagnosis of DUB was defined as the index date. We randomly matched a date between new diagnosis date of DUB and endpoint as the index date for the non-CHM cohort. Surgical events related to DUB, including hysterectomy (ICD-9-OP: 68.0, 68.3, 68.4, 68.41, 68.49, 68.5, 68.51, 68.59, 68.6, 68.7 and 69.0) and endometrial ablation (ICD-9-OP: 68.23) after the index date were measured. The outcome measurement was the comparison of incidences of surgical events in the two cohorts with the variable of comorbidities and drug used before the end of December 31, 2013. Statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, NC, U.S.A)., and p<0.05 in two-tailed tests indicate statistical significance. For each variable, we used Cox proportional hazard regression to analyze the hazard ratios (HRs) and 95% confidence interval (95% CI). The Kaplan-Meier method was used to determine the cumulative incidence of surgery in both cohorts, and the Log rank test was used to compare incidence curves between the CHM and non-CHM cohorts.

We matched 4642 randomly selected patients for each group by use of 1:1 propensity score analysis for the CHM and non-CHM users. The baseline characteristics of both groups are shown in Table 1 , and they were similar in age, comorbidities, and drugs used. CHM users were more dominant in the 18–39-year-old age group than the >40-year-old age group. Table 1 Characteristics of Dysfunctional Uterine Bleeding Patients According to Use of Chinese Herb Medicine after Matching Patients with DUB Variable Chinese Herb Medicine Used p -value* Non-CHM Users (n=4642) CHM Users (n=4642) n % n % Age Mean±SD (years) 34.51(10.21) 33.63(10.29) <0.0001 a Age Group, years 0.1656  18–39 3201 68.96 3211 69.17  40–64 1402 30.2 1407 30.31  Older than 65 39 0.84 24 0.52 Comorbidities  Inflammatory Disease of Ovary 964 20.77 902 19.43 0.1083  Anemia 449 9.67 449 9.67 0.99  Menorrhagia 285 6.14 277 5.97 0.7277  Dysmenorrhea 275 5.92 274 5.9 0.9649  Female infertility 158 3.4 144 3.1 0.4128  Hypotension 4 0.09 4 0.09 0.99 † Childbirth status 0.0396  No 2401 51.72 2500 53.86  Yes 2241 48.28 2142 46.14 Drug used  NSAID 4450 95.86 4600 99.1 <0.0001  Tranexamic Acid 2550 54.93 2986 64.33 <0.0001  Progesterone 2259 48.66 2735 58.92 <0.0001  Estrogen 1992 42.91 2310 49.76 <0.0001  Danazol 37 0.8 68 1.46 0.0023  GnRH agonists 2 0.04 2 0.04 0.99 † Interval Between Onset of DUB Disease and the Index Date, Days, Mean (Median) 952 (649) 915 (616) 0.0551 a Follow Time (Mean, Median) (Years) 5.17 (4.57) 6.30 (5.55) Note : *Chi-Square Test, a t -test, † fisher exact test. Abbreviations : DUB, dysfunctional uterine bleeding; CHM, Chinese herbal medicine; NSAID, nonsteroidal anti-inflammatory drug; GnRH, Gonadotropin Releasing Hormone. Characteristics of Dysfunctional Uterine Bleeding Patients According to Use of Chinese Herb Medicine after Matching Note : *Chi-Square Test, a t -test, † fisher exact test. Abbreviations : DUB, dysfunctional uterine bleeding; CHM, Chinese herbal medicine; NSAID, nonsteroidal anti-inflammatory drug; GnRH, Gonadotropin Releasing Hormone. The most common comorbidity was inflammatory disease of the ovary (almost 20%). Nearly all patients in both cohorts used NSAIDs, and approximately 65% of CHM users took ranexamic acid. The mean duration between the initial diagnosis of DUB and the first time receiving Chinese herbal medicine was approximately 915 days ( Table 1 ). During the follow-up period, there were 631 patients included in our study who received surgery ( Table 2 ). A higher incidence of surgery was revealed in the 40–64-year-old group (adjusted HR 2.41). Patients with comorbidities of inflammatory disease of the ovary, anemia, menorrhagia and dysmenorrhea were more likely to receive surgery than patients without comorbidities. Patients without childbirth status were more likely to receive surgery than the childbirth group. Overall, the incidence of surgery was significantly lower in the CHM cohort than in the non-CHM cohort (adjusted HR 0.27, 95% CI 0.22–0.33). Table 2 Cox Model with Hazard Ratios and 95% Confidence Intervals of Surgery Associated with Chinese Herb Medicine and Covariates among Dysfunctional Uterine Bleeding Patients Variable DUB Cohort Surgery no. (n=631) Crude* Adjusted † HR (95% CI) p-value HR (95% CI) p-value CHM Use  Non-CHM User 485 1.00 reference 1.00 reference  CHM User 146 0.27 (0.22–0.32) <0.0001 0.27 (0.22–0.33) <0.0001 Age Group, Years  18–39 275 1.00 reference 1.00 reference  40–64 352 3.05 (2.61–3.57) <0.0001 2.47 (2.05–2.99) <0.0001  Older than 65 4 1.51 (0.56–4.06) 0.4114 1.09 (0.40–2.96) 0.86 Comorbidities (ref=non-)  Inflammatory Disease of Ovary 162 1.45 (1.21–1.73) <0.0001 1.36 (1.13–1.63) 0.0012  Anemia 134 2.70 (2.23–3.27) <0.0001 2.01 (1.64–2.45) <0.0001  Menorrhagia 84 2.63 (2.09–3.31) <0.0001 1.93 (1.52–2.45) <0.0001  Dysmenorrhea 59 1.88 (1.44–2.46) <0.0001 1.80 (1.37–2.37) <0.0001  Female infertility 11 0.54 (0.3–0.97) 0.0408 0.63 (0.35–1.15) 0.131  Hypotension 1 1.92 (0.27–13.66) 0.5139 0.61 (0.08–4.39) 0.6206 Childbirth Status (ref=non-)  No 431 1.00 reference 1.00 reference  Yes 200 0.51 (0.43–0.60) <0.0001 0.82 (0.68–1.01) 0.0565 Drug used (ref=non-)  NSAID 566 0.12 (0.1–0.16) <0.0001 0.17 (0.13–0.22) <0.0001  Tranexamic Acid 383 0.96 (0.82–1.12) 0.5954 1.07 (0.91–1.27) 0.4063  Progesterone 300 0.71 (0.61–0.83) <0.0001 0.90 (0.76–1.07) 0.2402  Estrogen 271 0.80 (0.68–0.94) 0.0063 0.93 (0.78–1.10) 0.4021  Danazol 16 2.04 (1.24–3.35) 0.005 2.17 (1.31–3.59) 0.0026  GnRH Agonists 0 - - - - - - Notes : Crude HR * represented relative hazard ratio. Adjusted HR † represented adjusted hazard ratio: mutually adjusted for CHM use, age, comorbidities, childbirth status and drug used in Cox proportional hazard regression. Abbreviations : DUB, dysfunctional uterine bleeding; CHM, Chinese herbal medicine; NSAID, Nonsteroidal Anti-Inflammatory Drug; GnRH, Gonadotropin Releasing Hormone. Cox Model with Hazard Ratios and 95% Confidence Intervals of Surgery Associated with Chinese Herb Medicine and Covariates among Dysfunctional Uterine Bleeding Patients Notes : Crude HR * represented relative hazard ratio. Adjusted HR † represented adjusted hazard ratio: mutually adjusted for CHM use, age, comorbidities, childbirth status and drug used in Cox proportional hazard regression. Abbreviations : DUB, dysfunctional uterine bleeding; CHM, Chinese herbal medicine; NSAID, Nonsteroidal Anti-Inflammatory Drug; GnRH, Gonadotropin Releasing Hormone. The difference in the cumulative incidence of surgery between the two groups was illustrated through a Kaplan–Meier analysis ( Figure 2 ). The cumulative incidence of surgery was significantly lower in the CHM cohort than the non-CHM cohort group during the follow-up period (Log rank test, p < 0.001). A total of 146 patients in the CHM cohort (4.99 per 1000 person-years) and 485 patients in the non-CHM cohort (20.19 per 1000 person-years) received surgery (adjusted HR 0.27, 95% CI 0.22–0.33) ( Table 3 ). The incidence rates of surgery in the 18–39-year-old group and the 40–64-year-old group that used CHM were 3.36 and 8.83 per 1000 person years, respectively, which were lower than those in the comparison cohort (12.01 and 41.5 per 1000 person-years, respectively). In addition, the 18–39-year-old group and the 40–64-year-old group showed a 0.29-fold (95% CI: 0.22–0.39) and a 0.25-fold (95% CI: 0.2–0.33) lower risk of surgery, respectively, than the non-CHM cohort. CHM significantly decreased the risk of surgery in both the 18–39-year-old and the 40–64-year-old age groups. CHM also reduced the risk of surgery in DUB patients with or without comorbidities. Regardless of childbirth status or whether patients took NSAIDs, tranexamic acid or progesterone, fewer patients in the CHM cohort underwent surgery than in the non-CHM cohort. Table 3 Incidence Rates, Hazard Ratio and Confidence Intervals of Surgery for Dysfunctional Uterine Bleeding Patients with and without Chinese Herbal Medicine Treatment Stratified by Age, Comorbidities, Drug Used and Childbirth Status Variables Chinese Herb Medicine Used Compared with non-CHM User Non-CHM users CHM users Crude HR Adjusted HR (n=4642) (n=4642) Event Person Years IR † Event Person Years IR † (95% CI) (95% CI) Total 485 24,017 20.19 146 29,264 4.99 0.27(0.22–0.32)*** 0.27(0.22–0.33)*** Age Group  18–39 207 17,233 12.01 68 20,230 3.36 0.29(0.22–0.38)*** 0.29(0.22–0.39)***  40–64 274 6603 41.5 78 8832 8.83 0.25(0.19–0.32)*** 0.25(0.2–0.33)***  Older than 65 4 181 22.13 0 203 0 - - Co-Morbidities  No 251 16,365 15.34 77 20,046 3.84 0.27(0.21–0.34)*** 0.29(0.22–0.37)***  Yes 234 7652 30.58 69 9218 7.49 0.27(0.21–0.35)*** 0.26(0.2–0.34)*** Drug Used  No 40 165 243.09 3 71 42.23 0.22(0.07–0.72)* 0.17(0.05–0.68)*  Yes 445 23,852 18.66 143 29,193 4.9 0.28(0.23–0.34)*** 0.27(0.22–0.33)*** Childbirth Status  No 340 11,898 28.58 91 15,664 5.81 0.22(0.18–0.28)*** 0.23(0.18–0.29)***  Yes 145 12,119 11.96 55 13,600 4.04 0.35(0.26–0.48)*** 0.37(0.27–0.51)*** Notes : Adjusted HR† represented adjusted hazard ratio: mutually adjusted for CHM use, age, comorbidities, childbirth status and drug used in Cox proportional hazard regression. *:<0.05; *** p<0.001. Comorbidities and drug used list as listed in the methods section. Abbreviations : IR, incidence rates, per 1,000 person-years; CHM, Chinese herb medicine; HR, hazard ratio; CI, confidence interval. Figure 2 Cumulative incidence of surgery between the CHM cohort and non-CHM cohort. The cumulative incidence of surgeries related to dysfunctional uterine bleeding (DUB) in the CHM cohort (dashed line) is significantly lower than in the non-CHM cohort (solid line) (Log rank test, p<0.001). Incidence Rates, Hazard Ratio and Confidence Intervals of Surgery for Dysfunctional Uterine Bleeding Patients with and without Chinese Herbal Medicine Treatment Stratified by Age, Comorbidities, Drug Used and Childbirth Status Notes : Adjusted HR† represented adjusted hazard ratio: mutually adjusted for CHM use, age, comorbidities, childbirth status and drug used in Cox proportional hazard regression. *:<0.05; *** p<0.001. Comorbidities and drug used list as listed in the methods section. Abbreviations : IR, incidence rates, per 1,000 person-years; CHM, Chinese herb medicine; HR, hazard ratio; CI, confidence interval. Cumulative incidence of surgery between the CHM cohort and non-CHM cohort. The cumulative incidence of surgeries related to dysfunctional uterine bleeding (DUB) in the CHM cohort (dashed line) is significantly lower than in the non-CHM cohort (solid line) (Log rank test, p<0.001). In Tables 4 and 5 , the most commonly prescribed single herbs and multi-herbal products (formulas) for the treatment of patients with DUB are listed. Yi-Mu-Cao (Herba Leonuri; Leonurus heterophyllus Sweet) and Jia-Wei-Xiao-Yao-San (Bupleurum and Peony Formula) was the most commonly used single herb and formula, respectively. Table 4 The Most Commonly Prescribed Single Herbs for the Treatment of Dysfunctional Uterine Bleeding Pin-Yin Name Chinese Materia Medica name Botanical Name Indication for TCM syndrome Yi-Mu-Cao Herba Leonuri Leonurus heterophyllus Sweet Menstrual irregularities due to blood stasis with edema Xiang-Fu Rhizoma Cyperi Cyperus rotundus L. Irregular menstruation because of liver qi stagnation Dan-Shen Radix Salviae Miltiorrhizae Salvia miltiorrhiza Bge. Irregular menstruation because of blood and qi stagnation Yan-Hu-Suo Rhizoma Corydalis Corydalis yanhusuo W. T. Wang Irregular menstruation because of liver qi stagnation Xu-Duan Radix Dipsaci Dipsacus asperoides , C. Y.Chent et TM Ai Deficiency in liver and kidney Xian-He-Cao Herba Agrimoniae Agrimonia eupatoria L. var. pilosa Mak Excessive bleeding Han-Lian-Cao Herba ecliptae Eclipta prostrata  Linn, Eclipta alba  (L.) Hassk. Yin deficiency in liver and kidney; excessive bleeding during menopause Tu-Si-Zi Semen Cuscutae Chinensis Cuscuta chinensis Lam. Yin deficiency in liver and kidney; infertility Du-Zhong Cortex Eucommiae Ulmoidis Eucommia ulmoides Oliv. Deficiency in liver and kidney Nu-Zhen-Zi Fructus Ligustri Lucidi Ligustrum lucidum Yin deficiency in liver and kidney Abbreviation : TCM, Traditional Chinese medicine. Table 5 The Most Commonly Prescribed Chinese Herbal Formulas for the Treatment of Dysfunctional Uterine Bleeding Pin-Yin Name English Name Constitutions TCM Indication Pin-Yin Name Chinese Materia Medica name Botanical Name Jia-Wei-Xiao-Yao-San Bupleurum and Peony Formula Dang-Gui Fu-Ling Zhi-Zi Bo-He Bai-Shao Chai-Hu Gan-Cao Bai-Zhu Mu-Dan-Pi Wei-Jiang Radix Angelicae Sinensis Poria, Fructus Gardeniae Herba Menthae Haplocalycis Radix Paeoniae Alba Radix Bupleuri Radix Glycyrrhizae Rhizoma Atractylodis Macrocephalae Cortex Moutan Radicis Rhizoma Zingiberis officinales Angelica sinensis  (Oliv.) Diels Poria cocos  (Schw.) Wolf Gardenia jasminoides  J.Ellis Mentha haplocalyx  Briq.Field Paeonia lactiflora  Pall Bupleurum chinense  DC. Glycyrrhiza uralensis  Fisch Atractylis macrocephala Koidz Paeonia suffruticosa Andr. Zingiber officinale Rosc. Irregular menstruation because of spleen qi deficiency and liver blood deficiency with heat; liver qi stagnation Gui-Zhi- Fu-Ling-Wan Cinnamon and Poria Pills Gui-Zhi Fu-Ling Mu-Dan-Pi Chi-Shao Tao-Ren Ramulus Cinnamomi Cassiae Poria Cortex Moutan Radicis Radix Paeoniae Lactiflorae Semen Persicae Cinnamomum cassia Blume Poria cocos  (Schw.) Wolf Paeonia suffruticosa Andr. Paeonia lactiflora  Pall. Prunus persica  (L.) Batsch. Blood stasis in pelvic cavity Dang-Gui-Shao-Yao-San Tangkuei and Peony Powder Dang-Gui Fu-Ling Bai-Shao Bai-Zhu Chuan-Xiong Ze-Xie Radix Angelicae Sinensis Poria Radix Paeoniae Alba Rhizoma Atractylodis Macrocephalae Rhizoma Chuanxiong Rhizoma Alismatis Angelica sinensis  (Oliv.) Diels Poria cocos  (Schw.) Wolf Paeonia lactiflora Pall. Atractylodes macrocephala Koidz Ligusticum chuanxiong Hort. Alisma plantago-aquatica  L. Liver blood deficiency Wen-Jing-Tang Flow Warming Decoction Wu-Zhu-Yu Gui-Zhi Dang-Gui Chuan-Xiong Bai-Shao E-Jiao Mai-Men-Dong Mu-Dan-Pi Ren-Shen Gan-Cao Sheng-Jiang Ban-Xia Fructus Evodiae Rutaecarpae Ramulus Cinnamomi Cassiae Radix Angelicae Sinensis Radix Chuanxiong Radix Paeoniae Alba Colla Corii Asini Tuber Ophiopogonis Japonici Cortex Moutan Radicis Radix Ginseng Radix Glycyrrhizae Rhizoma Zingiberis officinales Rhizoma Pinelliae Ternatae Evodia rutaecarpa (Juss.) Benth. Cinnamomum cassia Blume Angelica sinensis  (Oliv.) Diels Ligusticum chuanxiong Hort. Paeonia lactiflora Pall. Equus asinus L. Ophiopogon japonicus (Thunb.) Ker_Gawl Paeonia suffruticosa Andr. Panax ginseng C. A. Mey Glycyrrhiza uralensis  Fisch Zingiber officinale Rosc. Pinellia ternate (Thunb.) Breit Pinellia   ternata  (Thunb.) Makino Blood stasis Xiong-Guei-Jiao-Ai-Tang Decoction of Donkey-Skin Glue and Artemisia Chuan-Xiong Dang-Gui E-Jiao Gan-Cao Shu-Di-Huang Bai-Shao Ai-Ye Rhizoma Chuanxiong Radix Angelicae Sinensis Colla Corii Asini Radix Glycyrrhizae Radix Rehmanniae Radix Paeoniae Alba Folium Artemisiae Argyi Ligusticum chuanxiong Hort. Angelica sinensis  (Oliv.) Diels Equus asinus L. Glycyrrhiza uralensis  Fisch Rehmannia glutinosa  (Gaertn.) Libosch. ex Fisch. and C.A. Mey. Paeonia lactiflora Pall Artemisia argyi  H.Lév. and Vaniot Blood deficiency; prevention of miscarriage Gui-Pi-Tang Restore the Spleen Decoction Ren-Shen Long-Yan-Rou Huang-Qi Gan-Cao Bai-Zhu Fu-Ling Mu-Xiang Dang-Gui Suan-Zao-Ran Yuan-Zhi Sheng-Jiang Da-Zao Radix Ginseng Arillus Euphoriae Longanae Radix Astragali Radix Glycyrrhizae Rhizoma Atractylodis Macrocephalae Poria Radix Aucklandiae Radix Angelicae Sinensis Semen Zizyphi Spinosae Radix Polygalae Tenuifoliae Radix Zingiberis officinalis Fructus Zizyphi Jujube Panax ginseng C. A. Mey Dimocarpus longan Lour. Astragalus henryi Oliv. Glycyrrhiza uralensis  Fisch Atractylis macrocephala Koidz Poria cocos  (Schw.) Wolf Aucklandia lappa Deene Angelica sinensis  (Oliv.) Diels Ziziphus jujube var. Spinosa (Bunge) Hu ex H. F. Chou Polygala tenuifolia Willd. Zingiber officinale Rosc. Ziziphus jujube Mill. Var. inermis Bge. Blood deficiency; vaginal spotting because of qi deficiency Shao-Fu-Zhu-Yu-Tang Drive Out Stasis from the Lower Abdomen Decoction Xiao-Hui-Xiang Pao-Jiang Yan-Hu-Suo Dang-Gui Chuan-Xiong Mo-Yao Rou-Gui Chi-Shao Pu-Huang Wu-Ling-Zhi Fructus Foeniculi Vulgaris Rhizoma Zingiberis officinales Rhizoma Corydalis Radix Angelicae Sinensis Rhizoma Chuanxiong Myrrh Ramulus Cinnamomi Cassiae Radix Paeoniae Lactiflorae Pollen Typhae Excrementum Trogopteri Xanthipes Foeniculum vulgare Mill. Zingiber officinale Rosc. Corydalis yanhusuo  W. T. Wang Angelica sinensis  (Oliv.) Diels Ligusticum chuanxiong Hort. Commiphora molmol Engl. Cinnamomum cassia Blume Paeonia lactiflora Pall. Typha angustifolia L. Trogopterus xanthipes Milne-Edwards Blood stasis and qi stagnation Tao-He-Cheng-Qi-Tang Kernel Qi- Coordinating Decoction Tao-Ren, Gui-Zhi Da-huang Mang-Xiao Zhi-Gan-Cao Semen Persicae Ramulus Cinnamomi Cassiae Radix et Rhizoma Rhei Natrii Sulfas, Radix Glycyrrhizae Prunus persica  (L.) Batsch Cinnamomum cassia Blume Rheum palmatum L. Mirabilitum Glycyrrhiza uralensis  Fisch Accumulation of blood stasis and heat in the lower burner Xiao-Yao-San Free Wanderer Powder Dang-Gui Fu-Ling Bo-He Bai-Shao Chai-Hu Gan-Cao Bai-Zhu Wei-Jiang Radix Angelicae Sinensis Poria Herba Menthae Haplocalycis Radix Paeoniae Alba Radix Bupleuri Radix Glycyrrhizae Rhizoma Atractylodis Macrocephalae Rhizoma Zingiberis officinales Angelica sinensis  (Oliv.) Diels Poria cocos  (Schw.) Wolf Mentha haplocalyx  Briq.Field Paeonia lactiflora  Pall Bupleurum chinense  DC. Glycyrrhiza uralensis  Fisch Atractylis macrocephala Koidz Zingiber officinale Rosc. Liver qi constraint with deficiency of the spleen Xue-Fu-Zhu-Yu-Tang House of Blood Stasis-Expelling Decoction Tao-Ren Hong-Hua Shu-Di-Huang Chi--Shao Dang-Gui Chuan-Xiong Niu-Xi Chai-Hu Jie- Geng Zhi-Ke Gan-Cao Semen Persicae Flos Carthami Tinctorii Radix Rehmanniae Radix Paeoniae Lactiflorae Radix Angelicae Sinensis Radix Ligustici Radix Achyranthis Bidentatae Radix Bupleuri, Radix Platycodonis Fructus CitriAurantii Radix Glycyrrhizae Prunus persica  (L.) Batsch Carthamus tinctorius  L. Rehmannia glutinosa Libosch Paeonia lactiflora Pall Angelica sinensis  (Oliv.) Diels Ligusticum chuanxiong Hort. Achyranthes bidentata  Bl. Bupleurum chinense  DC. Platycodon grandiflorus(Jacq.) A.DC. Citrus aurantium  L. Glycyrrhi za uralensis  Fisch Blood stasis in the “mansion of blood” Abbreviation : TCM, Traditional Chinese medicine. The Most Commonly Prescribed Single Herbs for the Treatment of Dysfunctional Uterine Bleeding Abbreviation : TCM, Traditional Chinese medicine. The Most Commonly Prescribed Chinese Herbal Formulas for the Treatment of Dysfunctional Uterine Bleeding Abbreviation : TCM, Traditional Chinese medicine.

Our study was the first nationwide population-based study investigating on the association of CHM with the need for surgeries (hysterectomy and endometrial ablation) in DUB patients. The results of our study demonstrated that integrative CHM treatment may be beneficial for patients with DUB. It is associated with a decreased incidence of surgery compared to the non-CHM group. The cumulative incidence rate of surgery in patients with DUB was significantly lower in CHM users in comparison with non-CHM users during the follow-up period. CHM users displayed an average 73% decreased risk of surgical events compared with non-CHM users (0.27, 95% CI 0.22–0.33, p40 years old have a higher incidence of dysfunctional uterine bleeding. Moreover, patients in that age group most likely no longer have plans for child-bearing. Our study revealed that younger patients preferred to receive CHM treatment, which was consistent with previous studies. 11 , 15 , 18 , 19 Patients who were taking danazol simultaneously were more likely to undergo surgery (adjusted HR 2.19, 95% CI 1.32–3.63), and patients who used NSAIDs were less likely to receive surgery (adjusted HR 0.17, 95% CI 0.13–0.22, p<0.001). Patients who took NSAIDs simultaneously tended to have better pain and bleeding control, 20 so the incidence of surgery was significantly reduced. Patients who used danazol were likely to have more complicated circumstances, 21 so the incidence of surgery was much higher. Patients with comorbidities of anemia, menorrhagia, dysmenorrhea and inflammatory disease of the ovary were more likely to have undergone hysterectomy or endometrial ablation. Patients with the comorbidities of dysmenorrhea and inflammatory disease of the ovary had pathological effects that were similar to those of DUB; these effects may have been caused by the dysregulated endometrial vascular development 22 and the imbalance of prostaglandins between the vasoconstriction effect and vasodilation of PGF2a, PGE2 and PGI2, which is the possible mechanism of DUB. 23–25 This is consistent with a previous study that found that dysmenorrhea may increase the risk of needing hysterectomy after endometrial ablation. 26 Pain or dysmenorrhea may be an important factor in the selection of DUB treatment. 27 , 28 Our study also revealed that patients with NSAID use had a lower incidence of surgery. The comorbidities of menorrhagia and associated anemia were consistent with the surgical treatment guidelines of DUB, which state that Hb<10 can increase the need for surgery. 29 The main disadvantage of medical treatment options is the potential recurrence of symptoms. Complications and long-term post-operation problems following hysterectomy cannot be ignored despite the fact that hysterectomy will immediately stop heavy menstrual bleeding. Hysterectomy impairs patients’ ovarian function and may cause early menopause, which in turn increases the risk for cardiovascular diseases. 30 Some studies have revealed an association between hysterectomy and high blood pressure, and an abnormal lipid profile. 31 , 32 Therefore, many women would likely benefit from trying TCM for help. We previously identified the prescription pattern of the CHM for patients with DUB. 11 From the perspective of TCM theory, these herbs can be categorized according to the TCM syndrome differentiation. Among the top 10 commonly prescribed single herbs and herbal formulas, some of them are used for relieving blood stasis and qi stagnation and thus might relieve pain, while some are used to nurture Yin and blood to restore the uterine function according to the TCM theory ( Table 4 and Table 5 ). It is necessary to clarify that these herbal prescriptions should be based on TCM diagnosis. The application of herbs may be individualized and thus the prescriptions may be different between patients. Yi-Mu-Cao (Herba Leonuri), Xiang-Fu (Rhizoma Cyperi) and Yan-Hu-Suo (Rhizoma Corydalis) were the mostly commonly used single herbs for patients with DUB. A couple experimental models may explain the potential mechanisms of these single herbs. Yi-Mu-Cao has been traditionally used to help activate blood and resolve stasis. Leonurine, an alkaloid present in Yi-Mu-Cao, has anti-fibrotic, antioxidant, anti-inflammatory and analgesic effects. 33 It can also induce and strengthen uterine contractions, so it is used in the treatment of menstrual disorders. 34 Additionally, Xiang-Fu (Rhizoma Cyperi) has been used as an estrogenic agent in estrogen-deprived mice. 35 The component tetrahydropalmatine from Yan-Hu-Suo was revealed to inhibit D2 dopamine receptors and have analgesic effects. 36 Our previous study also identified that Jia-Wei-Xiao-Yao-San, Dang-Gui-Shao-Yao-San and Gui-Zhi-Fu-Ling-Wan were commonly used for patients with DUB. 11 Jia-Wei-Xiao-Yao-San has been widely used to treat menstrual disorders and the emotional and psychological symptoms. It was reported that its anti-depressant effect might be related to regulation of TNF-α levels. 37 The pharmacological properties of Gui-Zhi-Fu-Ling-Wan have been reported to inhibit the cascade of the overproduction of COX-2 and iNOS. 38 It also significantly suppresses the protein and mRNA levels of MIF, IL-6, IL-8, and TNF- α . 39 It has been demonstrated to have protective effects against vascular injury and to inhibit the proliferation of uterine leiomyoma cells. 40 , 41 Dang-Gui-Shao-Yao-San significantly suppress oxytocin-evoked PGF2α production of rat endometrial epithelial cells and has antagonistic action on uterine contraction. 42 , 43 Clinically, the current treatment options for AUB can be medical, surgical, or a combination of both, depending on the underlying cause. It is also necessary to consider key factors such as the patient’s age, fertility desire, symptom relief, and co-morbidities. 44 Medical treatments typically involve iron supplementation and the use of hormonal or non-hormonal therapies. Surgical options include the removal of focal lesions, endometrial resection or destruction, and hysterectomy. 45 According to our study, CHM should be integrated as a kind of medical treatment for the preservation of fertility function, restoration of regular menstrual period, relieving symptoms and improving quality of life. The treatment should also be tailored according to the needs of the patients. For example, those who are at a child-bearing age should consider CHM treatment before undergoing surgery. There were some limitations to our research; for instance, the results of the laboratory data and the image examinations were not available in the database. Consequently, the disease severity between the TCM seekers and the non-TCM seekers could not be evaluated in this study. Thus, we tried to exclude those who were diagnosed as having cervical cancer, endometrial cancer, or ovarian cancer from our study population. We also used a propensity score to match the CHM and non-CHM cohorts. On the other hand, levonorgestrel intrauterine devices were not reimbursed by the NHI program until 2015, so we could not identify subjects who had used this device. Even though, our findings support the idea to conduct a high-quality randomized controlled trial to evaluate the efficacy and safety of Chinese herbal medicine.

This is the first large-scale population-based study on the surgery rate of TCM users among DUB patients. We found that complementary CHM is associated with a reduced surgical rate for DUB patients regardless of age, childbirth status, comorbidities, or drug use. Clinically, CHM should be integrated as a kind of medical treatment for the preservation of fertility function, restoration of regular menstrual period, relieving symptoms and improving quality of life. A high-quality randomized controlled clinical trial to determine the efficacy of CHM for DUB patients should be conducted in the future.