Journal of
Clinical Medicine Article
Investigation on the Use of Traditional Chinese Medicine for Polycystic Ovary Syndrome in a Nationwide Prescription Database in Taiwan Wan-Ting Liao 1,2 , Jen-Huai Chiang 3,4 , Chia-Jung Li 5 and Hung-Rong Yen 1,7,8,9,10,11, * ID 1 2 3 4 5 6 7 8 9 10 11
*
ID
, Ming-Tsung Lee 5 , Cheng-Chiung Su 6
Graduate Institute of Chinese Medicine, School of Chinese Medicine, College of Chinese Medicine, China Medical University, Taichung 404 Taiwan;
[email protected] Department of Chinese Medicine, Show Chwan Memorial Hospital, Changhua 500, Taiwan Management Office for Health Data, China Medical University Hospital, Taichung 404, Taiwan;
[email protected] College of Medicine, China Medical University, Taichung 404, Taiwan Research Assistant Center, Show Chwan Memorial Hospital, Changhua 500, Taiwan;
[email protected] (C.-J.L.);
[email protected] (M.-T.L.) Post Baccalaureate Medicine, Kaohsiung Medical University, Kaohsiung 807, Taiwan;
[email protected] Department of Chinese Medicine, China Medical University Hospital, Taichung 404, Taiwan Research Center for Traditional Chinese Medicine, Department of Medical Research, China Medical University Hospital, Taichung 404, Taiwan Research Center for Chinese Herbal Medicine, China Medical University, Taichung 404, Taiwan Chinese Medicine Research Center, China Medical University, Taichung 404, Taiwan Department of Biotechnology, Asia University, Taichung 413, Taiwan Correspondence:
[email protected]; Tel.: +886-4-2205-3366
Received: 28 June 2018; Accepted: 19 July 2018; Published: 22 July 2018
Abstract: Polycystic ovary syndrome (PCOS) is a common condition, affecting 5–10% of women of reproductive age worldwide. It has serious reproductive implications and causes mood disorders and metabolic disorders, such as type-2 diabetes. Because PCOS reflects multiple abnormalities, there is no single drug that can treat all its symptoms. Existing pharmaceutical agents, such as oral contraceptives (OCs), are suggested as a first-line therapy for menstrual irregularities; however, OCs are not appropriate for women pursuing pregnancy. Additionally, insulin-sensitizing agents, which appear to decrease insulin levels and hyperandrogenemia in women with PCOS, have been associated with a high incidence of gastrointestinal adverse effects. It is a common practice in Chinese society to receive traditional Chinese medicine (TCM) for treatment of gynecological problems and infertility. Current research demonstrates that several herbs and herbal formulas show beneficial effects in PCOS treatment. In this study, we conducted the first large-scale survey through the Taiwan National Health Insurance Program database to analyze TCM utilization patterns among women with PCOS in Taiwan during 1997–2010. The survey results revealed that 89.22% women with newly diagnosed PCOS had received TCM therapy. Jia-Wei-Xiao-Yao-San and Xiang-Fu (Rhizoma Cyperi) were the most commonly used formula and single herb, respectively, in the database. In addition, we found that the top five commonly prescribed single herbs and herbal formulas have shown promise in treating symptoms associated with PCOS. Keywords: Polycystic ovary syndrome; traditional Chinese medicine; Jia-Wei-Xiao-Yao-San
J. Clin. Med. 2018, 7, 179; doi:10.3390/jcm7070179
www.mdpi.com/journal/jcm
J. Clin. Med. 2018, 7, 179
2 of 15
1. Introduction Polycystic ovary syndrome (PCOS) is a common condition in humans, affecting 5–10% of women of reproductive age worldwide [1]. It has serious reproductive implications, such as anovulatory infertility, oligomenorrhea, amenorrhea, hyperandrogenism, and pregnancy complications [2]. It can also lead to mood disorders—according to a review report, the prevalence of depression among women with PCOS is 4 times greater than that among women without PCOS [3]. Research over the last few decades has revealed that PCOS is strongly associated with metabolic disorders, including an increased risk of insulin resistance, type-2 diabetes, obesity, and cardiovascular diseases (worsening of lipid profile and blood-vessel function; high blood pressure) [4]. Some reports have suggested that women with PCOS might have an increased prevalence of nonalcoholic fatty liver disease (NAFLD) and high C-reactive protein levels [5–7]. Because PCOS reflects multiple abnormalities, current treatment protocols for PCOS aim to achieve different goals, including healthy weight control, amelioration of hyperandrogenic symptoms, management of underlying metabolic and reproductive complications, and improvement of quality of life. In PCOS, weight management through lifestyle and behavioral intervention is a first-line treatment strategy recommended by evidence-based guidelines [8,9]. With regard to pharmaceutical treatments, oral contraceptives (OCs) are suggested as a first-line therapy for menstrual irregularities and hyperandrogenism [10,11]. However, OCs are not appropriate for women pursuing pregnancy, and they might cause weight gain, thus exacerbating the PCOS [12]. For fertility needs, clomiphene induces ovulation through the release of the gonadotropin-releasing hormone and, subsequently, gonadotropin from the anterior pituitary. However, pregnancy rates in clomiphene therapy have remained low among overweight women with PCOS [13–15]. Management approaches for metabolic disorders include treatment with metformin and thiazolidinediones, which appear to decrease insulin levels and hyperandrogenemia in women with PCOS [16]. However, use of metformin is associated with a high incidence of gastrointestinal adverse effects [17]. Moreover, because of the possibility of cardiovascular adverse events, thiazolidinediones are not suggested for nondiabetic women with PCOS [18,19]. It is a common practice in Chinese society to receive treatment with herbal remedies or acupuncture for obstetrical and gynecological conditions, such as uterine fibroids, dysmenorrhea, climacteric syndrome, premenstrual syndrome, and infertility. Previous studies have reported that, in women with PCOS, treatment with Chinese herbal medicine in conjunction with clomiphene citrate provides higher pregnancy rates than treatment with clomiphene citrate alone [20–23]. Moreover, acupuncture therapy might increase menstrual frequency and ovulation rates [24–28]. Research in diabetes and insulin resistance has shown progress in recent years. Integrating traditional Chinese medicine (TCM) healthcare into diabetes care has been shown to decrease the risk of developing kidney failure and stroke in type-2 diabetes mellitus [29,30]. Therefore, we hope to understand the clinical characteristics and treatment behavior of women with PCOS with regard to TCM. In Taiwan, TCM has been reimbursed under the current Taiwanese National Health Insurance (NHI) system since 1996. At the end of 2010, approximately 99.89% of the total 23 million population were enrolled in the NHI [31]. Although the broadly defined category of TCM includes Chinese herbal medicine, proprietary Chinese medicine, acupuncture, moxibustion, manipulation, mediation, Qi management, and other practices, the NHI program covers only three major modalities: (1) Chinese herbal products (CHPs) manufactured by GMP (good manufacturing practice)-certified pharmaceutical companies (e.g., concentrated scientific TCM granules and finished herbal products); (2) acupuncture, moxibustion, and cupping therapy; and (3) manipulative therapy (including acupressure, chiropractic, and tui na massage) [32]. All data regarding the use of TCM in Taiwan have been collected in the National Health Insurance Research Database (NHIRD). However, there is a lack of large-scale statistical data on the clinical usage of TCM in PCOS. Therefore, this study aimed to understand the TCM prescription patterns and combinations among women with PCOS in Taiwan in order to gain information for further pharmacologic experiments and clinical trials.
J. Clin. Med. 2018, 7, 179
3 of 15
2. Materials and Methods 2.1. Data Source This study used reimbursement claims data deposited in the NHIRD, which has been implemented by the Taiwan government since 1996. The NHIRD contains longitudinal data of a cohort comprising 1 million participants randomly selected from among insurance beneficiaries between January 1997 and December 2013 (LHID2000). These datasets include information on patient characteristics; date of clinic visit; diagnosis codes in accordance with the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) system; prescription details for Chinese herbal medicine; and detailed claims data for examination and disease management for all admitted patients and outpatients. In this study, NHRI datasets from 1997 to 2010 were selected for use as the research database, and patients were followed to December 2013. This study was approved by the institutional review board of China Medical University in central Taiwan (CMUH104-REC2-115). 2.2. Study Design and Population The flowchart for selection of PCOS cases is shown in Figure 1. This study included patients who were diagnosed with PCOS (ICD-9-CM code: 256.4) in NHIRD from 1 January 1997 to 31 December 2010. All cases had undergone gynecologic ultrasonography (19003C) or blood testing for testosterone or 17-hydroxyprogesterone levels (09121C, 09121B, or 09109C) within a year of diagnosis. We excluded the beneficiaries less than 18 years old (n = 508), if sex information indicated male, or if there was missing information for sex or birthday (n = 186). At last, participants who withdrew from the NHI were also excluded from our study (n = 20). Patients who had been receiving TCM between the date of initial diagnosis of PCOS and the study endpoint were included in the TCM users group (n = 5962). The first diagnosis date of accepted TCM after initial PCOS diagnosis date was considered the index date; patients with PCOS who had not taken any TCM after the initial PCOS diagnosis date were included in the non-TCM cohort (n = 720).
J. Clin. Med. 2018, 7, 179
4 of 15
J. Clin. Med. 2018, 7, x FOR PEER REVIEW
4 of 15
1,000,000 subject in LHID2000 file
All patients with PCOS who had undergone gynecologic ultrasonography (19003C) or blood test for testosterone or 17-hydroxyprogesterone levels (09121C, 09121B, or 09109C) in 1997 to 2010 yearsn=7396
Exclusion: 1. Less than 18 years (n=508). 2. Sex information is “male” or missing information for sex or birthday (n=186).
Individual with diagnosed PCOS patients in 1997 to 2010 n=6702
Exclusion: Participants withdrew from the NHI. (n=20)
Non-TCM user n=720
TCM user n =5962
Figure 1. 1. Flowchart Flowchart of of subjects subjects from from the the one one million million samples samples randomly randomly selected selected from from the the National National Figure Health Insurance Insurance Research Research Database Database (NHIRD). (NHIRD). TCM: TCM: traditional traditional Chinese Chinese medicine; medicine; PCOS: PCOS: polycystic polycystic Health ovary syndrome; NHI: National Health Insurance. ovary syndrome; NHI: National Health Insurance.
2.3. Potential Potential Confounders Confounders 2.3. We identified identified any any potential potential confounders confounders for for PCOS PCOS diagnosed diagnosed before before the the index index date, date, including including We the following following comorbidities: comorbidities: diabetes diabetes mellitus (ICD-9-CM: 250.x), 250.x), infertility infertility female female (628.x), (628.x), hirsutism hirsutism the mellitus (ICD-9-CM: (704.1), acne varioliformis (706.0 or 706.1), obesity (278), lipid metabolism disorders (272.0, (704.1), acne varioliformis (706.0 or 706.1), obesity (278), lipid metabolism disorders (272.0, 272.1, 272.1, 272.2, 272.2,or 272.3 or 272.4), major depression or 296.3x), and amenorrhea or 272.3 272.4), major depression (296.2x(296.2x or 296.3x), anxietyanxiety (300.x),(300.x), and amenorrhea (626.0 or(626.0 626.1). 626.1). 2.4. Statistical Analysis 2.4. Statistical Analysis Mean values and standard deviation were described for continuous variables, and percentages values standard deviation were described for continuous variables, andStudent percentages were Mean described forand categorical variables. Intergroup comparison was performed by the t-test were described for categorical variables. Intergroup comparison was performed by the Student and chi-square test for continuous and categorical variables, respectively. Statistical analysist-test was and chi-square andCary, categorical variables, respectively. analysis was performed usingtest SASfor 9.4 continuous (SAS Institute, NC, USA) software. Two-tailed pStatistical values < 0.05 indicated performed using SAS 9.4 (SAS Institute, Cary, NC, USA) software. Two-tailed p values < 0.05 statistical significance. indicated statistical significance. 3. Results 3. Results 3.1. Subject Characteristics 3.1. Subject Characteristics The present analysis included 6682 subjects who were newly diagnosed with PCOS during 1997–2010 (Table 1). All included enrolled 6682 patients had who undergone gynecologic ultrasonography or blood The present analysis subjects were newly diagnosed with PCOS during 1997– 2010 (Table 1). All enrolled patients had undergone gynecologic ultrasonography or blood testing for serum testosterone levels. Among the enrolled subjects, while 720 did not receive TCM for treatment
J. Clin. Med. 2018, 7, 179
5 of 15
testing for serum testosterone levels. Among the enrolled subjects, while 720 did not receive TCM for treatment of PCOS in the follow-up period, 5962 had received TCM for PCOS treatment. In both cohorts (TCM users and nonusers), the highest proportion of patients were in the age group of 18–29 years. There was no substantial difference in urbanization level or standard of hospitals (where PCOS was diagnosed) between the two groups. With regard to comorbidities, TCM users had a higher prevalence of infertility (p = 0.027), amenorrhea (p < 0.0001), and anxiety (p = 0.0002) than TCM nonusers. The prevalence of diabetes mellitus, obesity, lipid metabolism disorders, and major depression was similar in both groups. Table 1. Demographic characteristics of the patients newly diagnosed with PCOS in Taiwan in 1997–2010.
Variable
Without TCM
With TCM
n = 720 (10.78%)
n = 5962 (89.22%)
n
%
n
p Value
% 0.0654 *
Age at baseline 18–29 30–39 Older than 40 Mean(SD)
395 54.86 265 36.81 60 8.33 29.77(6.93)
3539 59.36 1991 33.39 432 7.25 29.09(6.67)
Urbanization 1 (highest) 2 3 4 4+ (lowest)
236 230 116 89 49
32.78 31.94 16.11 12.36 6.81
1972 1824 984 707 475
33.08 30.59 16.5 11.86 7.97
Hospital level (where diagnosed with PCOS) Medical center Regional hospital District hospital Physician clinics
170 281 118 150
23.64 39.08 16.41 20.86
1326 2172 1070 1394
22.24 36.43 17.95 23.38
Baseline comorbidity Diabetes mellitus Infertility female Hirsutism Acne varioliformis Obesity Disorders of lipoid metabolism Major depression Anxiety Amenorrhea
43 129 1 227 14 42 16 60 218
5.97 17.92 0.14 31.53 1.94 5.83 2.22 8.33 30.28
441 1362 27 1950 159 294 137 786 2630
7.4 22.84 0.45 32.71 2.67 4.93 2.3 13.18 44.11
0.0094 ‡ 0.7906 *
0.2209 *
Interval between onset of PCOS disease and the first TCM consultation, days mean (median) ‡
0.1636 * 0.0027 * 0.3567 * 0.5236 * 0.2489 * 0.2954 * 0.898 * 0.0002 * 6
Total of TCM (n = 5962)
3.3. TCM Prescription Patterns for PCOS We conducted an analysis to investigate the prescription patterns of CHPs and identified 10 of the most commonly prescribed Chinese herbal formulas and single herbs (Table 3). The most commonly prescribed Chinese herbal formula was Jia-Wei-Xiao-Yao-San (JWXYS; Supplemented Free Wanderer Powder), followed by Gui-Zhi-Fu-Ling-Wan (GZFLW; Cinnamon Twig and Poria Pill), and Dang-Gui-Shao-Yao-San (Lesser Abdomen Stasis-Expelling Decoction). The most commonly prescribed single herb was Xiang-Fu (Cyperus rotundus L.), followed by Da-Huang (Rheum officinale Baill.) and Yi-Mu-Cao (Leonurus artemisia (Lour.) S. Y. Hu). Table 3. Ten most common herbal formulas prescribed.
Herbal Formula
Frequency
Number of Person-Days
Average Daily Dose
Average Duration for Prescription
(g)
(Days)
Single Chinese herb Xiang-Fu (Cyperus rotundus L.) Da-Huang (Rheum officinale Baill.) Yi-Mu-Cao (Leonurus artemisia (Lour.) S. Y. Hu) Yan-hu-suo (Corydalis yanhusuo W. T. Wang) Dan-shen (Salvia miltiorrhiza Bge.) Gan-Cao (Glycyrrhiza uralensis Fisch.) Tu-Si-Zi (Cuscuta chinensis Lam.) Huang-Qin (Scutellaria baicalensis Georgi) Bei-Mu (Fritillaria thunbergii Miq.) Du-Zhong (Eucommia ulmoides Oliv.)
12,375 11,350 10,824 9889 8072 7818 7178 7616 5860 5414
86,491 83,520 77,687 68,526 61,438 53,026 52,268 52,213 39,858 39,649
1.0 0.7 1.1 1.1 1.1 0.9 1.3 1.2 1.1 1.1
7.0 7.4 7.2 6.9 7.6 6.8 7.3 6.9 6.8 7.3
Combined Chinese herb Jia-Wei-Xiao-Yao-San Gui-Zhi-Fu-Ling-Wan Dang-Gui-Shao-Yao-San Wen-Jing-Tang Ma-Zi-Ren-Wan Chuan-Xiong-Cha-Tiao-San Gui-Pi-Tang Ger-Gen-Tang Shao-Fu-ZhuYu-Tang Xue-Fu-Zhu-Yu-Tang
21,305 8779 8470 7970 6850 5643 4895 5471 4994 4473
15,3952 62,294 58,510 58,018 51,079 36,836 36,050 35,096 34,233 33,638
4.3 3.9 4.4 4.2 2.7 3.9 4.3 4.3 3.8 3.8
7.2 7.1 6.9 7.3 7.5 6.5 7.4 6.4 6.9 7.5
List according to person-days
3.4. Frequency Distribution of Disease Categories in TCM and Non-TCM Visits We analyzed the frequency distribution of disease categories for TCM and non-TCM visits on the basis of ICD-9-CM codes from the claims data. Among TCM users, genitourinary diseases (99.21%) were the most common reasons for visiting TCM clinics, followed by digestive (99.14%) and respiratory (99.03%) disorders. In case of TCM nonusers visiting Western medical clinics, digestive disorders (95.42%) were more frequent than genitourinary diseases (94.58) and respiratory disorders (94.44%). The disease patterns among TCM users who visited clinics were similar to those among TCM nonusers (Table 4).
J. Clin. Med. 2018, 7, 179
7 of 15
Table 4. The distribution of TCM and non-TCM users by major disease categories/diagnosis in patients with PCOS.
Disease (ICD-9-CM)
Without TCM (n = 720)
With TCM (n = 5962)
p Value
n
%
n
%
Infectious and parasitic disease (001–139)
439
60.97
4433
74.35
