Printer Friendly

Efficacy and Safety of Integrated Traditional Chinese Medicine and Western Medicine on the Treatment of Rheumatoid Arthritis: A Meta-Analysis.

1. Introduction

Rheumatoid arthritis (RA) is one of the most prevalent chronic systemic autoimmune diseases [1]. It is characterized by synovial membrane inflammation and hyperplasia, autoantibody production, cartilage and bone destruction, and systemic features [2]. The typical symptoms of RA are pain and swelling in the joints of hands and feet, accompanied by morning stiffness of the affected joints; large joints including shoulder, elbow, knee, and ankle joints could also be injured [3]. RA has a relatively constant incidence of 0.5% to 1% [4], and population-based epidemiologic studies consistently reveal that family history of RA increases the risk of the onset of it by 3-5 times [5].

The therapeutic targets of RA are focusing on reducing joint inflammation and pain, maximizing joint function, and preventing from articular destruction and deformity. Treatment regimens are composed of medications, weight-bearing exercise, health education, and rest [6]. Western medicine (WM) treatment for RA mainly includes nonsteroidal anti-inflammatory drugs (NSAIDs), disease-modifying antirheumatic drugs (DMARDs), glucocorticoids (GC), and biological agents. NSAIDs are not used to control the disease progression of RA [7]. Methotrexate (MTX) among DMARDs is still the first-line choice for treating RA recommended by the international guidelines [8], but researches indicated generally low remission rates with MTX monotherapy [9]. GC is the most potent disease-modifying drug in clinic at present, but its chronic use could cause osteoporosis, osteonecrosis, and other hazards [10]. Biological agents are expensive and their long-term effects are still controversial, though they have a positive effect on symptom reduction of RA [11, 12].

China has abundant botanical resources which have been widely used in RA treatment [13-15]. Tripterygium wilfordii Hook. f., Aconitum carmichaelii Debx., and Curcuma longa L. represent a few of the many medicines of botanical origin for RA in traditional Chinese medicine (TCM), which may have a positive effect not only on the symptoms but also on the disease progression [16-18]. Formula is the main category of herbal remedies. Guizhi-Shaoyao-Zhimu Decoction is a representative prescribed formula to treat RA. A synthetic approach [19] that combined drug target prediction, network analysis, and experimental validation indicated that Guizhi-Shaoyao-Zhimu Decoction may partially attenuate RA by means of reversing inflammation-immune system disequilibrium and regulating the HDAC1-HSP90AA1-NFKB2-IKBKB-TNF-[alpha] signaling axis. As one of novel Chinese patent medicines, Xinfeng capsule shows benefits in alleviating joint pain, swelling, and early morning stiffness, and it could also ameliorate extra-articular manifestations such as anemia, platelet disorder, lipid metabolism disturbance, abnormal cardiopulmonary function, depression, and quality of life with few adverse reaction [20, 21]. Many effective ingredients of antirheumatic Chinese herbs have been found to inhibit RA development and some of the effective extracts have been verified. Luo et al. [22] summarized evidences on the efficacy and safety of clinical application of tripterygium glycosides and total glucosides of paeony, suggesting that they might be potential beneficial complementary and alternative medicines for RA patients. Artemisia asiatica has a long history of ethnopharmacological use in Asian countries such as China, Korea, and Japan, and a novel antioxidative and anti-inflammatory formulation prepared from the ethanol extracts of Artemisia asiatica named DA-9601 is now on sale in South Korea [23, 24]. A recent study [25] has shown that DA-9601 injection reduced arthritis scores in collagen-induced arthritis mice; moreover, eupatilin, the main active component of DA-9601, could markedly downregulate the expression of inflammatory cytokines and suppress the differentiation of osteoclasts, indicating that DA-9601 and eupatilin are candidate anti-inflammatory agents.

TCM has special superiorities in reducing the adverse reactions of WM and improving its curative effect [26, 27]. So, the combination of TCM and conventional WM provides a new approach for the improvement of quality of life and disease control of RA patients. Many studies showed that the integrated TCM-WM therapy has a positive effect on the treatment of RA. However, due to the small sizes of multisamples and uneven quality of articles, it is difficult to draw reliable conclusions based on small-sample randomized controlled trials (RCTs). Therefore, we conducted this meta-analysis aiming to systematically evaluate the efficacy and safety of integrated TCM-WM versus WM mono-therapy for the treatment of RA. We supposed that this research could provide the evidence for the superiority of treating RA with integrative medicine.

2. Methods

2.1. Search Strategy. Associated studies from inception to January 2020 were retrieved in the following electronic databases: PubMed, EMBASE, and the Cochrane Library. The search strategies for each database are presented in the Supplementary file 1. In addition, the reference lists of relevant publications were manually searched to find additional studies. The searches were independently performed by two authors.

2.2. Inclusion and Exclusion Criteria. The following were included: (1) studies published in either English or Chinese language; (2) participating patients diagnosed with RA in accordance with the 1987 American Rheumatism Association (1987 ARA) or the 2010 American College of Rheumatology and European Union League Against Rheumatism (2010 ACR/EULAR) diagnostic criteria; (3) experimental groups (EGs) treated with a combination of TCM and WM, while control groups (CGs) treated only with WM; (4) RCTs; and (5) detailed data of at least 1 relevant outcome.

The following were excluded: (1) participants not diagnosed with RA according to the diagnostic criteria mentioned above; (2) participants restricted to special crowd (e.g., the elderly and juveniles); (3) EGs treated only with TCM; (4) duplicative data; (5) incomplete or unavailable data; and (6) reviews, conference abstracts, and case reports.

2.3. Types of Outcome Measures. The primary outcomes analyzed in this meta-analysis were therapeutic effects (TEs) and adverse events (AEs). The secondary outcomes were tender joint count (TJC), swollen joint count (SJC), duration of morning stiffness (DMS), grip strength (GS), disease activity score in 28 joints (DAS28), rheumatoid factor (RF), anti-cyclic peptide containing citrulline (anti-CCP), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP).

TEs were associated with the improvements of clinical symptoms and laboratory indexes, and the most used remission criterion was ACR20/50/70 [28]. ACR20 signified 20% improvements in TJC and SJC as well as 20% improvements in at least 3 of the 5 following items: (1) patient assessment of pain; (2) patient global assessment of disease activity; (3) physician global assessment of disease activity; (4) health assessment questionnaire (HAQ); and (5) acute-phase reactants (ESR and CRP). ACR50 and ACR70 represented 50% and 70% improvements, respectively. The response to treatments was evaluated excellent if the overall improvement of ACR70 was 70%; good if the overall improvement of ACR50 was between 50% and 69%; moderate if the overall improvement of ACR20 was between 20% and 49%; and poor if the treatment did not meet the ACR20 standard. TEs were calculated from the number of excellent, good, and moderate results.

All data were acquired directly from the original studies. Dichotomous variables (TEs and AEs) were expressed as absolute numbers, and continuous data (TJC, SJC, DMS, GS, DAS28, RF, anti-CCP, ESR, and CRP) were expressed as mean with standard deviation for further analysis.

2.4. Data Extraction and Quality Assessment. The relevant data were selected and extracted independently by two authors, including names of authors, publication years, sample sizes, ages, genders, courses of the disease, intervention methods, durations of intervention, and outcome indexes. Disagreements were resolved by discussing with a third investigator.

The qualities of the studies included were evaluated by each author on the basis of the Cochrane collaboration's tool [29] for bias risk assessing. The assessments were performed on the following: (1) random sampling method; (2) allocation concealment method; (3) blinding of subjects and experimenters; (4) blinding of outcome assessment; (5) the completion of outcome data; (6) report selection; and (7) other bias, such as specific research designs that could affect the overall outcomes. The results of the 7 items above were assessed as low risk, unclear, or high risk.

2.5. Statistical Analysis. All included studies were analyzed with Review Manager 5.3 software (The Cochrane Collaboration, Copenhagen, Denmark). Odds ratios (OR) and 95% confidence intervals (CI) were calculated for dichotomous data, while mean differences (MD), standardized mean differences (SMD), and 95% CI were calculated for continuous data. Heterogeneity was statistically assessed using the chi-squared test and the [I.sup.2] statistic, and [I.sup.2] > 50% indicated obvious heterogeneity among trials [30]. The analysis was carried out by the use of a random-effect model if P < 0.1 or [I.sup.2] > 50% but a fixed-effect model if P [greater than or equal to] 0.1 or [I.sup.2] [less than or equal to] 50%. Descriptive approaches would be adopted if the data were insufficient. Publication bias was detected using funnel plot.

3. Results

3.1. Study Search and Selection. Initially 364 publications were identified, including 67 articles from PubMed, 134 articles from EMBASE, 163 articles from the Cochrane Library, and no record from manual search. After exclusion of duplicates, 221 studies were screened. Through further evaluation, 20 studies (Wu et al. [31]; Lu et al. [32]; Zhao and Liu [33]; Liu et al. [34]; Li et al. [35]; Lin et al. [36]; Zhao and Wang [37]; Huang et al. [38]; Yu and Yu [39]; Wang et al. [40]; Chen et al. [41]; Wang and Tao [42]; Qian et al. [43]; Jiang et al. [44]; Zhang et al. [45]; Wang [46]; Chen [47]; Du et al. [48]; Yang et al. [49]; and Huang et al. [50]) met the predefined inclusion criteria and were included in this meta-analysis. The general procedure for study selection is summarized in Figure 1.

3.2. Description of the Studies. Table 1 summarized the characteristics of the 20 included trials which were studies. There were a total of 2269 cases over all included studies, 2214 (1192 cases for oral TCM combined with WM and 1022 cases for oral WM alone) of them completed the studies. The studies were published between 2001 and 2019, and they were all carried out in China. Sixteen studies were published in Chinese, while 4 studies were in English. Fifteen studies [31-43, 47, 48] used the 1987 ARA diagnostic criteria, including 2 studies [38, 48] combined with the 2010 ACR/ EULAR criteria, while 5 studies [44-46, 49, 50] used the 2010 ACR/EULAR criteria. All of the RCTs demonstrated no significant difference in baseline characteristics between experimental and control groups. Of these RCTs, the study population of Huang et al. [38] comprised 28 male patients and 52 female patients with mean age of 36.8 [+ or -] 9.3 years and mean disease course of 3.7 [+ or -] 2.3 years; the study of Chen et al. [41] comprised 31 male and 165 female participants with mean age of 44.6 [+ or -] 13.3 years, including those who had severe adverse reactions and withdrew their consents. The interventions were limited to Chinese herbal medicine and the conventional WM. TCM used in these studies included Chinese herbal decoctions and tablets or capsules made from herbs such as Qingbi Tablet, Kunxian Capsule, and Xinfeng Capsule, or their extracts such as tripterygium glycosides, total glucosides of paeony, sinomenine and the extract of Artemisia annua L. WM included DMARDs, NSAIDs, and GC, and the most common of these was MTX. The groups treated with WM only were considered to be the control groups. The foremost outcomes of the included studies were TEs, and all of these studies described them. Eleven studies mentioned the TJC with 10 [31, 32, 34, 36, 38-40, 42, 49, 50] conforming to the desired form of data; 11 mentioned the SJC with 10 [31, 32, 34, 36, 38-40, 42, 49, 50] meeting requirements; 9 mentioned the DMS with 9 [31-33, 36, 38-40, 42, 50] meeting requirements; 6 mentioned the GS with 5 [31-33, 39, 40] meeting requirements; 8 mentioned the DAS28 with 5 [38, 40, 44, 48, 50] meeting requirements; 13 reported the effects on RF with 12 [31-33, 36-39, 42, 44, 45, 48, 49] meeting requirements; 8 reported the effects on anti-CCP with 7 [33, 36, 37, 44, 45, 48, 49] meeting requirements; 17 studies reported the effects on ESR with 15 [31-34, 36-40, 42, 44, 45, 48-50] meeting requirements; and 15 reported the effects on CRP with 14 [32-34, 36-40, 42, 44, 45, 48-50] meeting requirements. In addition, 18 [31-38, 40-46, 48-50] of these studies discussed the AEs in detail.

3.3. Risk of Bias Assessment. A summary of the risks of bias in the 20 studies included in the meta-analysis is presented in Figures 2 and 3. For most of the items in the included trials, the risks of bias were low or unclear. All the studies included were described as RCTs; among them, 13 studies [32, 34, 35, 37, 40, 42-46, 48-50] adequately represented the random methods. Allocation concealment and blinding methods were poorly reported. Only 2 trials [35, 50] mentioned allocation concealment methods; others did not specify whether allocation concealment was performed, so the risks of bias in allocation concealment of them were unknown. Two trials [41, 50] were open-label with high risks in performance and detection biases. Regarding incomplete data, which is attrition bias, the authors judged that there was no missing data or that the reasons for the missing outcome indicators could not possibly be related to the true value of the outcomes. Since original study protocols and adequate relevant information were not available to assess selective reporting, all trials were considered to have unclear risks in reporting bias. Five trials [34, 40, 44, 45, 48] were judged at high risk in other bias, for they only focused on specific syndrome types or disease stages of RA, while the others were at low risk.

3.4. Effects of Interventions

3.41. Clinical Therapeutic Efficacy. All of the studies demonstrated TEs of the integrated TCM-WM compared with WM only for RA. There was no significant heterogeneity ([I.sup.2] = 0%, P = 0.77). Therefore, the analysis used a fixed-effect model. The outcome indicated that TEs in the experimental group were significantly better than in the control group (OR = 3.03, 95% CI [2.36, 3.88], P < 0.00001) (Figure 4).

3.4.2. Clinical Symptoms. Ten trials provided available TJC data with 682 cases in the experimental group and 559 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sub.2] = 92%, P < 0.00001). A significant difference was discovered in TJC between 2 groups (MD = -1.17, 95% CI [-2.12, -0.21], P = 0.02), as shown in Figure 5.

Ten trials provided available SJC data with 682 cases in the experimental group and 559 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 96%, P < 0.00001). A significant difference was discovered in SJC between 2 groups (MD = -0.87, 95% CI [-1.85, 0.10], P = 0.08), as shown in Figure 6.

Nine trials provided available DMS data with 593 cases in the experimental group and 481 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 79%, P < 0.00001). A significant difference was discovered in DMS between 2 groups (SMD = -0.69, 95% CI [-0.98, -0.41], P < 0.00001), as shown in Figure 7.

Five trials provided available GS data with 355 cases in the experimental group and 275 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 94%, P < 0.00001). A significant difference was discovered in GS between 2 groups (SMD = 0.12, 95% CI [-0.63, 0.87], P = 0.75), as shown in Figure 8.

Five trials provided available DAS28 data with 318 cases in the experimental group and 285 cases in the control group, and a fixed-effect model was conducted to analyze the data ([I.sup.2] = 48%, P = 0.10). A significant difference was discovered in DAS28 between 2 groups (MD = -0.43, 95% CI [-0.57, -0.29], P < 0.00001), as shown in Figure 9.

3.4.3. Laboratory Indexes. Twelve trials provided available RF data with 655 cases in the experimental group and 525 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 85%, P < 0.00001). A significant difference was discovered in RF between 2 groups (SMD = -0.59,95% CI [-0.91, -0.27], P = 0.0003), as shown in Figure 10.

Seven trials provided available anti-CCP data with 373 cases in the experimental group and 329 cases in the control group, and a fixed-effect model was conducted to analyze the data (I2 = 0%, P = 0.74). A significant difference was discovered in anti-CCP between 2 groups (SMD = -0.21, 95% CI [-0.36, -0.06], P = 0.006), as shown in Figure 11.

Fifteen trials provided available ESR data with 907 cases in the experimental group and 751 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 91%, P < 0.00001). A significant difference was discovered in ESR between 2 groups (MD = -8.36, 95% CI [-12.60, -4.12], P = 0.0001), as shown in Figure 12.

Fourteen trials provided available CRP data with 872 cases in the experimental group and 716 cases in the control group, and a random-effect model was conducted to analyze the data ([I.sup.2] = 97%, P < 0.00001). A significant difference was discovered in CRP between 2 groups (MD = -6.73, 95% CI [-9.38, -4.08], P < 0.00001), as shown in Figure 13.

3.4.4. Adverse Drug Reactions. AEs caused by combined TCM-WM or WM alone were reported in 18 of the studies. The most common AEs in both groups were gastrointestinal disorders, abnormal liver function, leukopenia, skin allergies and rashes, headaches and dizziness, and alopecia. Most of the studies were not affected by these AEs; only 6 studies [35, 36, 44, 45, 48-50] reported that some participants withdrew from the trials because of serious AEs. No heterogeneity was identified among the trials ([I.sup.2] = 0%, P = 0.99) based on a fixed-effect model. As shown in Figure 14, a statistically significant difference was presented between the overall AEs in 2 groups. According to the meta-analysis, the experimental group had fewer AEs than the control group.

3.5. Funnel Plot. TEs were used to measure publication bias. Funnel plot was conducted based on all of studies included (Figure 15). The results revealed that the funnel plot was graphic symmetrical in general and the patterns were concentrated in the middle-upper part except for 3 offsets, which indicated a mild publication bias.

4. Discussion

RA is a common internal medical disease mainly affected by both environmental and genetic factors [51]. If not treated promptly, it may lead to joint deformity or even complete loss of joint function, thus affecting the daily activities and working abilities of patients, and have high disability and teratogenic rate [52]. WM treatment plays a role in relieving inflammation, reducing pain, and slowing joint damage; though the overall effects are positive, there are deficiencies, such as more adverse reactions and expensive costs, that ought to by no means be ignored. In recent years, there has been an increase in the use of integrated TCM-WM to treat RA. The integrative medicine combines the advantages of the theoretical experience of TCM with conventional WM, aiming to increase the efficacy, minimize adverse reactions during treatment, and improve prognosis of the diseases.

Early diagnosis and treatment are likely to influence the outcomes of the disease and even the remission conditions [53]. Autoantibodies RF and anti-CCP belong to the seral biomarkers involved in the 2010 ACR/EULAR RA classification criteria, exhibiting essential sero-diagnostic utility [54]. Combination of indicators of RF and anti-CCP makes for specific diagnosis of RA [55]. Acute-phase reactants ESR and CRP are important means for assessing the degree of activity of chronic inflammatory lesions as the increases in the levels of these clinical inflammatory markers indicate high disease activity [56]. DAS28 was reported in a mass of daily practice as well as clinical trials in RA [57]. The most common composite index of remission employs the DAS using 44 or 28 joint counts; the latter goes by the name of DAS28, which could monitor the disease evolution. DAS28 < 2.6 is generally considered to be in remission [58]. Therefore, in addition to clinical symptoms and adverse reactions, this meta-analysis was also used to evaluate the effects of integrated TCM-WM on the regulation of these indexes in RA patients, by which providing evidence-based medical basis for the clinical application of integrated medicine. Compared with WM alone, the combination of TCM and WM treatment could increase TEs, and the improvements of TJC, SJC, DMS, GS, DAS28, RF, anti-CCP, ESR, and CRP values were prominent in this study.

Some related findings might provide explanations for the therapeutic effects of integrated TCM-WM treatment in RA. Li et al. [59] pointed out that abnormal cellular immunity, such as high percentages of peripheral blood [CD4.sup.+], [CD8.sup.+], and [CD4.sup.+]/[CD8.sup.+] ratio, and increased IgG and IgA levels existed in RA patients. After 1 month of integrated TCM-WM treatment, the [CD4.sup.+]/[CD8.sup.+] ratio and the levels of IgG and IgA decreased obviously, demonstrating that combination of TCM and WM could regulate the balance of T-lymphocyte subsets. Other researchers [60] chose RA patients with damp-heat-obstruction symptom pattern as research subjects and divided them into TCM Sanhuangyilong decoction plus MTX group and MTX-only group. It was found that TNF-[alpha] and IFN-[gamma] may play a part in the development of RA. After 4 weeks of treatments, TNF-[alpha] and IFN-[gamma] levels were significantly decreased in Sanhuangyilong decoction plus MTX group, and the differences in TNF-[alpha] and IFN-[gamma] between 2 groups were statistically significant. Moreover, the combined treatment had more clinical benefits than MTX only. Liu et al. [61] compared the treatment characteristics of TCM and WM on the articular cartilage erosion related biochemical and immune factors and found that TCM mainly increased red blood cell count which bounded up with the degree of cartilage damage while platelet count decreased after WM treatment, showing that both TCM and WM could ameliorate cartilage damage in RA, but acted in different ways.

Drug-drug interactions have always been an active area that cannot be ignored in clinical medicine research. Some drugs can be used in combination to obtain an effectiveness that cannot be achieved with TCM or WM alone, but some may cause AEs and even endanger life. As the main means of treating RA, WM may cause a variety of AEs, especially gastrointestinal disorders, abnormal liver function, leukopenia, skin allergies and rashes, headaches and dizziness, and alopecia, which could affect patient compliance to some extent. In contrast, the frequency and severity of AEs in the treatment of integrated medicine were lower than those in WM in this meta-analysis. However, in order to ensure safe medication, we had better continue paying attention to this area. In Taiwan, a multi-TCM/WM interactions database was built to report the prevalence of interactions between TCM and WM, which could issue timely alerts when embedded inside the hospital clinical information system [62].

This study has several strengths: first, since the study included not only Chinese trials but also English trials, we have obtained a greater range of data than any other previous study in China. Moreover, the study did not limit patients to specific TCM or WM treatment options, which means that the results could be applied more extensively to RA patients. Furthermore, we collected as many outcome indicators as possible to acquire a more comprehensive evaluation of the effectiveness of treatment. Still, we are supposed to consider the following limitations: (1) all included studies were conducted in China, so there was a certain racial bias; (2) most of them had no or just a brief description of the principle of randomization, allocation concealment, or blinding; and (3) the TCM or WM regimens involved in the various studies were not entirely consistent, and there were also differences in the dosage and course of treatment under the same regimens, which increased statistical difficulty. Heterogeneities were found in some outcome indicators of this study, which could influence the accuracy and reliability of the results. Correctness of data was first checked to confirm that heterogeneities were not caused by data entry errors. Due to factors such as small sample sizes, loose experimental designs, different treatment durations, and inconsistent interventions, the outcomes were affected to varying degrees, which may also result in certain heterogeneity of results. In order to obtain reliable meta-analysis results, this study used the strategy of changing the statistical effect model. Based on the above, we recommend the following changes in clinical studies: (1) larger sample sizes, multiple centers, and longer follow-up times are required; (2) strict inclusion and exclusion criteria should be developed and outcome assessment and safety analysis need to be standardized; (3) randomization, assignment of concealment, blinding, and other information should be described, and patients who lost follow-up or dropped out of the studies are supposed to be recorded timely, thereby reducing methodological heterogeneity and reporting bias, and further improving the quality of evidence-based medicine research.

After a systematic review of 20 articles with 2269 cases, the study found that comprehensive medical treatment of RA has been widely proved to be therapeutic. Compared with WM, integrated treatment of RA is a more preferable intervening measure, with obvious advantages in improving efficacy and reducing adverse reactions. Nevertheless, prospective, large-sample, and long-term trials are needed in the future.

5. Conclusion

This meta-analysis demonstrated the possibility that the combination of TCM and WM for the treatment of RA might be more effective and safer than WM monotherapy. In addition to effectively improving clinical symptoms and reducing laboratory indexes, it may cause fewer side effects. Therefore, we suggest that integrated TCM-WM could be applied to the clinical treatment of RA. Further researches should aim to standardize RA treatment in order to strengthen the basis for combining TCM with WM.

https://doi.org/10.1155/2020/4348709

Conflicts of Interest

The authors declare no conflicts of interest regarding this work.

Acknowledgments

This research was supported by the National Natural Science Foundation of China (no. 81973741).

Supplementary Materials

Supplementary file 1: Search strategies. (Supplementary Materials)

References

[1] J. S. Smolen, D. Aletaha, and I. B. McInnes, "Rheumatoid arthritis," The Lancet, vol. 388, no. 10055, pp. 2023-2038, 2016.

[2] I. B. McInnes and G. Schett, "The pathogenesis of rheumatoid arthritis," New England Journal of Medicine, vol. 365, no. 23, pp. 2205-2219, 2011.

[3] D. Aletaha and J. S. Smolen, "Diagnosis and management of rheumatoid arthritis," JAMA, vol. 320, no. 13, pp. 1360-1372, 2018.

[4] A. J. Silman and J. E. Pearson, "Epidemiology and genetics of rheumatoid arthritis," Arthritis Research & Therapy, vol. 4, no. 3, pp. S265-S272, 2002.

[5] X. Jiang, T. Frisell, J. Askling et al., "To what extent is the familial risk of rheumatoid arthritis explained by established rheumatoid arthritis risk factors?" Arthritis & Rheumatology, vol. 67, no. 2, pp. 352-362, 2015.

[6] J. Bullock, S. A. A. Rizvi, A. M. Saleh et al., "Rheumatoid arthritis: a brief overview of the treatment," Medical Principles and Practice, vol. 27, no. 6, pp. 501-507, 2018.

[7] S. Trelle, S. Reichenbach, S. Wandel et al., "Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis," BMJ, vol. 342, no. 1, p. c7086, 2011.

[8] J. S. Smolen, R. Landewe, J. Bijlsma et al., "EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2016 update," Annals of the Rheumatic Diseases, vol. 76, no. 6, pp. 960-977, 2017.

[9] K. Chatzidionysiou and P. P. Sfikakis, "Low rates of remission with methotrexate monotherapy in rheumatoid arthritis: review of randomised controlled trials could point towards a paradigm shift," RMD Open, vol. 5, no. 2, Article ID e000993, 2019.

[10] M. C. Van Der Goes, J. W. G. Jacobs, M. Boers et al., "Monitoring adverse events of low-dose glucocorticoid therapy: EULAR recommendations for clinical trials and daily practice," Annals of the Rheumatic Diseases, vol. 69, no. 11, pp. 1913-1919, 2010.

[11] J. Listing, A. Strangfeld, S. Kary et al., "Infections in patients with rheumatoid arthritis treated with biologic agents," Arthritis & Rheumatism, vol. 52, no. 11, pp. 3403-3412, 2005.

[12] R. J. Looney, R. Srinivasan, and L. H. Calabrese, "The effects of rituximab on immunocompetency in patients with autoimmune disease," Arthritis & Rheumatism, vol. 58, no. 1, pp. 5-14, 2008.

[13] S. Lu, Q. Wang, G. Li, S. Sun, Y. Guo, and H. Kuang, "The treatment of rheumatoid arthritis using Chinese medicinal plants: from pharmacology to potential molecular mechanisms," Journal of Ethnopharmacology, vol. 176, pp. 177-206, 2015.

[14] K. L. Soeken, S. A. Miller, and E. Ernst, "Herbal medicines for the treatment of rheumatoid arthritis: a systematic review," Rheumatology, vol. 42, no. 5, pp. 652-659, 2003.

[15] S. Ahmed, J. Anuntiyo, C. J. Malemud, and T. M. Haqqi, "Biological basis for the use of botanicals in osteoarthritis and rheumatoid arthritis: a review," Evidence-Based Complementary and Alternative Medicine, vol. 2, no. 3, pp. 301-308, 2005.

[16] J. Wang, N. Chen, L. Fang et al., "A systematic review about the efficacy and safety of Tripterygium wilfordii Hook.f. preparations used for the management of rheumatoid arthritis," Evidence-Based Complementary and Alternative Medicine, vol. 2018, Article ID 1567463, 13 pages, 2018.

[17] Y. N. He, S. P. Ou, X. Xiong et al., "Stems and leaves of Aconitum carmichaelii Debx. as potential herbal resources for treating rheumatoid arthritis: chemical analysis, toxicity and activity evaluation," Chinese Journal of Natural Medicines, vol. 16, no. 9, pp. 644-652, 2018.

[18] D. Dorin, G. Marilena, G. Laura et al., "Phytomedicine in joint disorders," Nutrients, vol. 9, no. 1, p. 70, 2017.

[19] Q. Guo, X. Mao, Y. Zhang et al., "Guizhi-shaoyao-zhimu decoction attenuates rheumatoid arthritis partially by reversing inflammation-immune system imbalance," Journal of Translational Medicine, vol. 14, no. 1, p. 165, 2016.

[20] J. Liu, Y. Wang, C. Huang et al., "Efficacy and safety of xinfeng capsule in patients with rheumatoid arthritis: a multi-center parallel-group double-blind randomized controlled trial," Journal of Traditional Chinese Medicine, vol. 35, no. 5, pp. 487-498, 2015.

[21] J. Liu and R. L. Liu, "The potential role of Chinese medicine in ameliorating extra-articular manifestations of rheumatoid arthritis," Chinese Journal of Integrative Medicine, vol. 17, no. 10, pp. 735-737, 2011.

[22] J. Luo, W. J. Song, Y. Xu, G. Y. Chen, Q. Hu, and Q. W. Tao, "Benefits and safety of tripterygium glycosides and total glucosides of paeony for rheumatoid arthritis: an overview of systematic reviews," Chinese Journal of Integrative Medicine, vol. 25, no. 9, pp. 696-703, 2019.

[23] A. Ahuja, Y. S. Yi, M. Y. Kim, and J. Y. Cho, "Ethno-pharmacological properties of Artemisia asiatica : a comprehensive review," Journal of Ethnopharmacology, vol. 220, pp. 117-128, 2018.

[24] S. C. Choi, E. J. Choi, H. M Oh et al., "DA-9601, a standardized extract of Artemisia asiatica, blocks TNF-alpha-induced IL-8 and CCL20 production by inhibiting p38 kinase and NF-kappaB pathways in human gastric epithelial cells," World Journal of Gastroenterology, vol. 12, no. 30, pp. 4850-4858, 2006.

[25] J. Kim, Y. Kim, H. Yi et al., "Eupatilin ameliorates collagen induced arthritis," Journal of Korean Medical Science, vol. 30, no. 3, pp. 233-239, 2015.

[26] X. Wang, Y. Zu, L. Huang et al., "Treatment of rheumatoid arthritis with combination of methotrexate and Tripterygium wilfordii : a meta-analysis," Life Sciences, vol. 171, pp. 45-50, 2017.

[27] N. Xiang, X. M. Li, M. J. Zhang et al., "Total glucosides of paeony can reduce the hepatotoxicity caused by methotrexate and leflunomide combination treatment of active rheumatoid arthritis," International Immunopharmacology, vol. 28, no. 1, pp. 802-807, 2015.

[28] V. K. Ranganath, D. Khanna, and H. E. Paulus, "ACR remission criteria and response criteria," Clinical and Experimental Rheumatology, vol. 24, no. 43, pp. 14-21, 2006.

[29] J. P. T. Higgins, D. G. Altman, P. C. Gotzsche et al., "The cochrane collaboration's tool for assessing risk of bias in randomised trials," BMJ, vol. 343, p. d5928, 2011.

[30] J. P. T. Higgins, S. G. Thompson, J. J. Deeks et al., "Measuring inconsistency in meta-analyses," BMJ, vol. 327, no. 7414, pp. 557-560, 2003.

[31] Y. J. Wu, Z. Y. Lao, and Z. L. Zhang, "Clinical observation on small doses Tripterygium wilfordii polyglycoside combined with methotrexate in treating rheumatoid arthritis," Chinese Journal of Integrated Traditional and Western Medicine, vol. 21, no. 12, pp. 895-896, 2001.

[32] S. J. Lu, J. Shao, and X. R. Li, "Clinical observation on treatment of rheumatoid arthritis by combined therapy with methotrexate, sulfasalazine and Chinese herbal medicine," Chinese Journal of Integrated Traditional and Western Medicine, vol. 22, no. 8, pp. 571-573, 2002.

[33] Y. X. Zhao and Y. Liu, "Clinical observation on effects of leflunomid and total glucosides of paeony on rheumatoid arthritis," Chinese Journal of Integrated Traditional and Western Medicine, vol. 26, no. 4, pp. 355-357, 2006.

[34] W. Liu, X. Y. Liu, and Y. Wang, "Effect of Chinese herbs in enhancing prednisone for treatment of refractory rheumatoid arthritis," Chinese Journal of Integrated Traditional and Western Medicine, vol. 27, no. 8, pp. 742-744, 2007.

[35] E. K. Li, L. S. Tam, C. K. Wong et al., "Safety and efficacy of Ganoderma lucidum (lingzhi) and san miao san supplementation in patients with rheumatoid arthritis: a double-blind, randomized, placebo-controlled pilot trial," Arthritis & Rheumatism, vol. 57, no. 7, pp. 1143-1150, 2007.

[36] C. S. Lin, X. Y. Yang, and L. Dai, "Multi-center clinical study on therapeutic effect of kunxian capsule on rheumatoid arthritis," Chinese Journal of Integrated Traditional and Western Medicine, vol. 31, no. 6, pp. 769-774, 2011.

[37] S. S. Zhao and J. Wang, "Short-term clinical observation on compound xiatianwu combined with methotrexate in treating rheumatoid arthritis," Zhongguo Zhong Yao Za Zhi, vol. 37, no. 23, pp. 3664-3666, 2012.

[38] C. B. Huang, J. Liu, X. Chen et al., "Treatment of rheumatoid arthritis by xinfeng capsule: an efficacy observation," Chinese Journal of Integrated Traditional and Western Medicine, vol. 33, no. 12, pp. 1599-1602, 2013.

[39] S. Y. Yu and Z. A. Yu, "Clinical observation of compound xiatianwu tablets in treatment of 120 cases with active rheumatoid arthritis," Zhongguo Zhong Yao Za Zhi, vol. 38, no. 6, pp. 899-901, 2013.

[40] J. M. Wang, Q. W. Tao, Y. Z. Zhang et al., "Treating rheumatoid arthritis patients of Shen deficiency and cold invading syndrome by bushen quhan zhiwang decoction combined methotrexate: an evaluation of clinical efficacy and safety," Chinese Journal of Integrated Traditional and Western Medicine, vol. 33, no. 5, pp. 614-618, 2013.

[41] Z. Chen, X. P. Li, Z. J. Li, L. Xu, and X. M. Li, "Reduced hepatotoxicity by total glucosides of paeony in combination treatment with leflunomide and methotrexate for patients with active rheumatoid arthritis," International Immuno-pharmacology, vol. 15, no. 3, pp. 474-477, 2013.

[42] Z. Wang and X. J. Tao, "Treatment of rheumatoid arthritis by yangxue tongluo recipe combined with immunosuppressive agents: a clinical observation," Chinese Journal of Integrated Traditional and Western Medicine, vol. 34, no. 3, pp. 276-278, 2014.

[43] X. Qian, Z. L. Sun, G. Wei et al., "Clinical efficacy and safety of Chinese medicine decoction combined with western medicine in treatment of rheumatoid arthritis," Liaoning Journal of Traditional Chinese Medicine, vol. 42, no. 12, pp. 2371-2373, 2015.

[44] P. Jiang, L. Y. Zhang, L. L. Dai et al., "Treatment of rheumatoid arthritis by hebi recipe: an efficacy observation," Chinese Journal of Integrated Traditional and Western Medicine, vol. 36, no. 1, pp. 24-28, 2016.

[45] Y. Y. Zhang, J. Wang, P. Jiang et al., "Efficacy of hebi formula combined methotrexate on early rheumatoid arthritis patients with dis- harmony of gan and pi syndrome and its effects on serum MMP-3 and RANK/RANKL/OPG expressions," Chinese Journal of Integrated Traditional and Western Medicine, vol. 36, no. 10, pp. 1197-1201, 2016.

[46] A. Y. Wang, "Clinical study on treatment of rheumatoid arthritis with traditional Chinese medicine and western medicine," China Foreign Medical Treatment, vol. 9, no. 3, pp. 175-176, 2016.

[47] X. Z. Chen, "Clinical retrospective analysis of 80 cases of rheumatoid arthritis treated with integrated Chinese and western medicine," Chinese Journal of Communication, vol. 32, no. 5, pp. 108-109, 2016.

[48] S. G. Du, Z. H. Guo, Q. Q. Kong et al., "Effect of sanbitang recipe in treatment of rheumatoid arthritis with kidney empty and cold-dampness symptom," Zhongguo Zhong Yao Za Zhi, vol. 42, no. 14, pp. 2802-2807, 2017.

[49] M. Yang, M. Y. Guo, Y. Luo et al., "Effect of Artemisia annua extract on treating active rheumatoid arthritis: a randomized controlled trial," Chinese Journal of Integrative Medicine, vol. 23, no. 7, pp. 496-503, 2017.

[50] R. Y. Huang, H. D. Pan, J. Q. Wu et al., "Comparison of combination therapy with methotrexate and sinomenine or leflunomide for active rheumatoid arthritis: a randomized controlled clinical trial," Phytomedicine, vol. 57, pp. 403-410, 2019.

[51] C. Croia, R. Bursi, D. Sutera et al., "One year in review 2019: pathogenesis of rheumatoid arthritis," Clinical and Experimental Rheumatology, vol. 37, no. 3, pp. 347-357, 2019.

[52] Y. Zhou, X. Wang, Y. An et al., "Disability and health-related quality of life in Chinese patients with rheumatoid arthritis: a cross-sectional study," International Journal of Rheumatic Diseases, vol. 21, no. 9, pp. 1709-1715, 2018.

[53] M. Schneider and K. Kruger, "Rheumatoid arthritis--early diagnosis and disease management," Deutsches Arzteblatt International, vol. 110, no. 27-28, pp. 477-484, 2013.

[54] B. I. Gavrila, C. Ciofu, and V. Stoica, "Biomarkers in rheumatoid arthritis, what is new?" Journal of Medicine and Life, vol. 9, no. 2, pp. 144-148, 2016.

[55] S. Rantapaa-Dahlqvist, B. A. W. De Jong, E. Berglin et al., "Antibodies against cyclic citrullinated peptide and IgA rheumatoid factor predict the development of rheumatoid arthritis," Arthritis & Rheumatism, vol. 48, no. 10, pp. 2741-2749, 2003.

[56] B. Heidari, K. Hajian, and A. R. Firous Jahi, "Correlation between serum CRP levels and disease activity in rheumatoid arthritis," Kowsar Medical Journal, vol. 9, pp. 208-220, 2004.

[57] P. L. Van Riel and J. Fransen, "DAS28: a useful instrument to monitor infliximab treatment in patients with rheumatoid arthritis," Arthritis Research & Therapy, vol. 7, no. 5, pp. 189-190, 2005.

[58] J. Fransen, M. C. Creemers, and P. L. Van Riel, "Remission in rheumatoid arthritis: agreement of the disease activity score (DAS28) with the ARA preliminary remission criteria," Rheumatology, vol. 43, no. 10, pp. 1252-1255, 2004.

[59] Y. N. Li, Y. S. Zhao, and X. Li, "Study on abnormality and regulation of T-lymphocyte subsets in peripheral blood of rheumatoid arthritis patients," Chinese Journal of Integrated Traditional and Western Medicine, vol. 22, no. 5, pp. 359-361, 2002.

[60] D. F. Liu, J. Yan, M. D. Yun et al., "Effect of sanhuangyilong decoction plus methotrexate on tumor necrosis factor alpha and interferon gamma in serum and synovial fluid in rheumatoid arthritis patients with symptom pattern of damp heat obstruction," Journal of Traditional Chinese Medicine, vol. 36, no. 5, pp. 625-633, 2016.

[61] X. W. Liu, Q. L. Zha, Y. T. He et al., "Comparative study on characteristics of Chinese and western medicine for treatment of rheumatoid arthritis regarding cartilage erosion related blood biochemical and immunological factors," Chinese Journal of Integrated Traditional and Western Medicine, vol. 27, no. 12, pp. 1090-1093, 2007.

[62] K. C. Chen, R. Lu, U. Iqbal et al., "Interactions between traditional Chinese medicine and western drugs in Taiwan: a population-based study," Computer Methods and Programs in Biomedicine, vol. 122, no. 3, pp. 462-470, 2015.

Qi Xing [ID], Ling Fu, Zhichao Yu, and Xueping Zhou [ID]

The First Clinical Medical College, Nanjing University of Chinese Medicine, Nanjing 210023, China

Correspondence should be addressed to Xueping Zhou; zhouxp1960@163.com

Received 23 October 2019; Revised 26 January 2020; Accepted 20 February 2020; Published 2 April 2020

Academic Editor: Alfredo Vannacci

Caption: Figure 1: Flow diagram of study selection process.

Caption: Figure 2: Risk of bias graph.

Caption: Figure 15: Funnel plot.
Table 1: Study characteristics.

                                      Sample size
                                     (male/female)
Author                 Year
                                   EG             CG

Wu et al. [31]         2001     35 (8/27)      35 (7/28)

Lu et al. [32]         2002     40 (9/31)      20 (5/15)

Zhao and Liu [33]      2006    40 (18/22)     40 (14/26)

Liu et al. [34]        2007    60 (12/48)     60 (10/50)

Li et al. [35]         2007     32 (5/27)      33 (4/29)

Lin et al. [36]        2011    79 (19/60)     79 (13/66)

Zhao and Wang [37]     2012     64 (8/56)      40 (4/36)

Huang et al. [38]      2013        40             40

Yu and Yu [39]         2013    120 (38/82)    60 (18/42)

Wang et al. [40]       2013    120 (31/89)    120 (33/87)

Chen et al. [41]       2013        105            89

Wang et al. [42]       2014     47 (8/39)      41 (6/35)

Qian et al. [43]       2015    84 (32/52)     84 (33/51)

Jiang et al. [44]      2016     32 (3/29)      31 (4/27)

Zhang et al. [45]      2016        36             36

Wang [46]              2016    28 (10/18)      28 (9/19)

Chen [47]              2016    40 (11/29)     40 (12/28)

Du et al. [48]         2017    56 (20/36)     56 (21/35)

Yang et al. [49]       2017    79 (16/63)     80 (18/62)

Huang et al. [50]      2019    73 (18/55)      47 (7/40)

                                       Age (years)
Author
                                EG                     CG

Wu et al. [31]          58.6 [+ or -] 2.6      56.7 [+ or -] 1.8

Lu et al. [32]          41.5 [+ or -] 11.2     40.6 [+ or -] 13.2

Zhao and Liu [33]       31.0 [+ or -] 8.9      30.0 [+ or -] 9.6

Liu et al. [34]        44.13 [+ or -] 19.29   43.75 [+ or -] 14.52

Li et al. [35]            50 [+ or -] 10         50 [+ or -] 13

Lin et al. [36]        51.76 [+ or -] 11.67   48.62 [+ or -] 13.01

Zhao and Wang [37]      42.4 [+ or -] 12.6     40.7 [+ or -] 11.1

Huang et al. [38]               --                     --

Yu and Yu [39]          37.1 [+ or -] 11.5     36.5 [+ or -] 10.4

Wang et al. [40]       31.62 [+ or -] 14.28   33.93 [+ or -] 12.46

Chen et al. [41]                --                     --

Wang et al. [42]       42.82 [+ or -] 12.45   44.78 [+ or -] 12.38

Qian et al. [43]                43                     45

Jiang et al. [44]         41 [+ or -] 10         43 [+ or -] 10

Zhang et al. [45]       42.0 [+ or -] 9.6      43.1 [+ or -] 9.5

Wang [46]               35.5 [+ or -] 6.6      35.9 [+ or -] 6.9

Chen [47]               37.2 [+ or -] 14.6     37.6 [+ or -] 11.9

Du et al. [48]         33.47 [+ or -] 12.37   36.52 [+ or -] 14.57

Yang et al. [49]       47.59 [+ or -] 14.43   44.70 [+ or -] 16.41

Huang et al. [50]      48.97 [+ or -] 10.79   48.53 [+ or -] 12.10

                                    Disease course
Author
                               EG                    CG

Wu et al. [31]         42.5 [+ or -] 15.1    40.0 [+ or -] 11.9
                             months                months

Lu et al. [32]          2.6 [+ or -] 1.2      2.8 [+ or -] 1.3
                             years                 years

Zhao and Liu [33]       4.0 [+ or -] 3.8      5.0 [+ or -] 4.9
                             years                 years

Liu et al. [34]        10.5 [+ or -] 7.64    9.63 [+ or -] 7.57
                             years                 years

Li et al. [35]          9.3 (4.8, 18.0)       7.8 (5.5, 11.5)
                             years                 years

Lin et al. [36]        5.46 [+ or -] 6.11    5.03 [+ or -] 4.24
                             years                 years

Zhao and Wang [37]      2.2 [+ or -] 0.6      2.0 [+ or -] 0.5
                             years                 years

Huang et al. [38]              --                    --

Yu and Yu [39]          2.9 [+ or -] 1.2      2.8 [+ or -] 1.2
                             years                 years

Wang et al. [40]       6.56 [+ or -] 4.63    7.17 [+ or -] 5.82
                             years                 years

Chen et al. [41]               --                    --

Wang et al. [42]        3.8 [+ or -] 6.2      4.0 [+ or -] 6.4
                             years                 years

Qian et al. [43]         0.3-146 months        0.8-142 months

Jiang et al. [44]       5.6 [+ or -] 1.6      5.8 [+ or -] 1.9
                             months                months

Zhang et al. [45]       5.6 [+ or -] 1.6      5.8 [+ or -] 1.9
                             months                months

Wang [46]               4.7 [+ or -] 2.5      4.9 [+ or -] 2.8
                             years                 years

Chen [47]               6.6 [+ or -] 2.6      6.3 [+ or -] 3.1
                             years                 years

Du et al. [48]         7.12 [+ or -] 3.72    6.93 [+ or -] 4.13
                             years                 years

Yang et al. [49]       6.46 [+ or -] 6.92    7.18 [+ or -] 8.37
                             years                 years

Huang et al. [50]        32.99 [+ or -]     40.56 [+ or -] 54.52
                          44.21 months             months

                                    Intervention
Author
                              EG                  CG

Wu et al. [31]            TWP 10 mg,        MTX 15 mg, qw +
                           tid + MTX            NSAIDs
                            7.5 mg,
                          qw + NSAIDs

Lu et al. [32]            FS1 30 ml,          PLB 30 ml,
                           bid + MTX           bid +MTX
                           5~10 mg,            5~10 mg,
                           qw + SSZ            qw + SSZ
                          0.5~1.0 g,          0.50~1.0 g,
                         tid + NSAIDs        tid + NSAIDs

Zhao and Liu [33]         TGP 0.6 g,         LEF 10 mg, qd
                       tid + LEF 10 mg,
                              qd

Liu et al. [34]           QT 5pills,        Voltaren 75 mg,
                           tid +PDN         qd + HCQ 0.2 g,
                                               qd + MTX
                                             5~15 mg, im,
                                                qw +PDN

Li et al. [35]            CPM 1.5 g,          PLB 1.5 g,
                          bid +DMARDs         bid +DMARDs
                         or NSAIDs or        or NSAIDs or
                              PDN                 PDN

Lin et al. [36]         KXC 0.3~0.6 g,       MTX 10 mg, qw
                           tid +MTX
                           10 mg, qw

Zhao and Wang [37]        CPM 0.6 g,         MTX 10 mg, qw
                           tid +MTX
                           10 mg, qw

Huang et al. [38]          XC 1.5 g,         MTX 10 mg, qw
                           tid + MTX
                           10 mg, qw

Yu and Yu [39]            CPM 0.6 g,          LEF 20 mg,
                       tid + LEF 20 mg,     qd + SSZ 1.0 g,
                        qd + SSZ 1.0 g,     bid + celecoxib
                        bid + celecoxib       0.2 g, bid
                          0.2 g, bid

Wang et al. [40]         BQZD 200 ml,        MTX 10 mg, qw
                           bid +MTX
                           10 mg, qw

Chen et al. [41]          TGP 0.6 g,          MTX 10 mg,
                           tid +MTX         qw + LEF 20 mg,
                        10 mg, qw + LEF           qd
                           20 mg, qd

Wang et al. [42]        YTR 0.5 agent,        MTX 10 mg,
                           bid + MTX        qw + LEF 10 mg,
                        10 mg, qw + LEF           qd
                           10 mg, qd

Qian et al. [43]         CHD 1 agent,        MTX 10 mg, qw
                           qd + MTX
                           10 mg, qw

Jiang et al. [44]         HR 150 ml,              MTX
                           bid + MTX         7.5~12.5 mg,
                         7.5-12.5 mg,       qw + folic acid
                        qw + folic acid    tablet 10 mg, qw
                       tablet 10 mg, qw

Zhang et al. [45]         HF 150 ml,              MTX
                           bid + MTX         7.5~12.5 mg,
                         7.5~12.5 mg,       qw + folic acid
                        qw + folic acid    tablet 10 mg, qw
                       tablet 10 mg, qw

Wang [46]               CHD 0.5 agent,        MTX 10 mg,
                           bid + MTX        qw + LEF 10 mg,
                        10 mg, qw + LEF           qd
                           10 mg, qd

Chen [47]                   CHD+MTX         MTX 7.5~20 mg,
                         7.5~20 mg, qw            qw

Du et al. [48]            CHD 300 ml,         MTX 7.5 mg,
                           bid + MTX              biw
                          7.5 mg, biw

Yang et al. [49]           EAA 30 g,          LEF 10 mg,
                        qd + LEF 10 mg,        qd + MTX
                           qd + MTX          7.5~15 mg, qw
                         7.5~15 mg, qw

Huang et al. [50]         SIN 120 mg,         LEF 20 mg,
                           bid + MTX           qd + MTX
                           10~15 mg,           10~15 mg,
                        qw + folic acid     qw + folic acid
                       tablet 5 mg, bid/   tablet 5 mg, bid/
                              tid                 tid

Author                 Duration               Outcomes

Wu et al. [31]         3 months      (1)(2)(3)(4)(5)(7)(9)(10)

Lu et al. [32]         24 weeks    (1)(2)(3)(4)(5)(7)(9)(10)(11)

Zhao and Liu [33]      12 weeks      (1)(4)(5)(7)(8)(9)(10)(11)

Liu et al. [34]        20 weeks         (1)(2)(3)(9)(10)(11)

Li et al. [35]         24 weeks         (1)(2)(3)(9)(10)(11)

Lin et al. [36]        12 weeks   (1)(2)(3)(4)(5)(6)(7)(8)(9)(11)

Zhao and Wang [37]     3 months         (1)(7)(8)(9)(10)(11)

Huang et al. [38]      12 weeks    (1)(2)(3)(4)(6)(7)(9)(10)(11)

Yu and Yu [39]         3 months      (1)(2)(3)(4)(5)(7)(9)(10)

Wang et al. [40]       24 weeks    (1)(2)(3)(4)(5)(6)(9)(10)(11)

Chen et al. [41]       24 weeks       (1)(6)(7)(8)(9)(10)(11)

Wang et al. [42]       12 weeks      (1)(2)(3)(4)(7)(9)(10)(11)

Qian et al. [43]       1 month                (1)(11)

Jiang et al. [44]      24 weeks       (1)(6)(7)(8)(9)(10)(11)

Zhang et al. [45]      24 weeks         (1)(7)(8)(9)(10)(11)

Wang [46]              12 weeks               (1)(11)

Chen [47]              3 months                 (1)

Du et al. [48]         16 weeks       (1)(6)(7)(8)(9)(10)(11)

Yang et al. [49]       48 weeks    (1)(2)(3)(6)(7)(8)(9)(10)(11)

Huang et al. [50]      24 weeks      (1)(2)(3)(4)(6)(9)(10)(11)

Quantitative data are shown as mean [+ or -] standard deviation or
median (interquartile range) or range. EG: experimental group; CG:
control group; TWP: Tripterygium wilfordii polyglycoside; MTX:
methotrexate; NSAIDs: nonsteroidal anti-inflammatory drugs; FS1:
Fengshi no.1; SSZ: sulfasalazine; PLB: placebo; TGP: total
glucosides of paeony; LEF: leflunomide; QT: Qingbi Tablet; PDN:
prednisone; HCQ: hydroxychloroquine; CPM: Chinese patent medicine;
DMARDs: disease-modifying antirheumatic drugs; KXC: Kunxian
Capsule; XC: Xinfeng Capsule; BQZD: Bushen Quhan Zhiwang Decoction;
YTR: Yangxue Tongluo Recipe; CHD: Chinese herbal decoction; HR:
Hebi Recipe; HF: Hebi Formula; EAA: the extract of Artemisia annua
L.; SIN: sinomenine; (1): therapeutic effects (TEs); (2): tender
joint count (TJC); (3): swollen joint count (STC); (4): duration of
morning stiffness (DMS); (5): grip strength (GS); (6): disease
activity score in 28 joints (DAS28); (7): rheumatoid factor (RF);
(8): anti-cyclic peptide containing citrulline (anti-CCP); (9):
erythrocyte sedimentation rate (ESR); (10): C-reactive protein
(CRP); (11): adverse events (AEs).

Figure 3: Risk of bias summary.

                  Zhao YX 2006   Zhao SS 2012   Zhang YY 2016

Random sequence       (b)            (a)             (a)
generation
(selection
bias)

Allocation            (b)            (b)             (b)
concealment
(selection
bias)

Blinding of           (b)            (b)             (b)
participants
and personnel
(performance
bias)

Blinding of           (b)            (b)             (b)
outcome
assessment
(detection
bias)

Incomplete            (a)            (a)             (a)
outcome data
(attrition
bias)

Selective             (b)            (b)             (b)
reporting
(reporting
bias)

Other bias            (a)            (a)             (c)

                  Yu SY 2013   Yang M 2017   Wu YJ 2001   Wang Z 2014

Random sequence      (b)           (a)          (b)           (a)
generation
(selection
bias)

Allocation           (b)           (a)          (b)           (b)
concealment
(selection
bias)

Blinding of          (b)           (b)          (b)           (b)
participants
and personnel
(performance
bias)

Blinding of          (b)           (b)          (b)           (b)
outcome
assessment
(detection
bias)

Incomplete           (a)           (a)          (a)           (a)
outcome data
(attrition
bias)

Selective            (b)           (b)          (b)           (b)
reporting
(reporting
bias)

Other bias           (a)           (a)          (a)           (a)

                  Wang JM 2013   Wang AY 2016   Qian X 2015

Random sequence       (a)            (a)            (a)
generation
(selection
bias)

Allocation            (b)            (b)            (b)
concealment
(selection
bias)

Blinding of           (b)            (b)            (b)
participants
and personnel
(performance
bias)

Blinding of           (b)            (b)            (b)
outcome
assessment
(detection
bias)

Incomplete            (a)            (a)            (a)
outcome data
(attrition
bias)

Selective             (b)            (b)            (b)
reporting
(reporting
bias)

Other bias            (c)            (a)            (a)

                  Lu SJ 2002   Liu W 2007   Lin CS 2011   Li EK 2007

Random sequence      (a)          (a)           (b)          (a)
generation
(selection
bias)

Allocation           (b)          (b)           (b)          (a)
concealment
(selection
bias)

Blinding of          (a)          (b)           (b)          (a)
participants
and personnel
(performance
bias)

Blinding of          (b)          (b)           (b)          (a)
outcome
assessment
(detection
bias)

Incomplete           (a)          (a)           (a)          (a)
outcome data
(attrition
bias)

Selective            (b)          (b)           (b)          (b)
reporting
(reporting
bias)

Other bias           (a)          (c)           (a)          (a)

                  Jiang P 2016   Huang RY 2019   Huang CB 2013

Random sequence       (a)             (a)             (b)
generation
(selection
bias)

Allocation            (b)             (a)             (b)
concealment
(selection
bias)

Blinding of           (b)             (c)             (b)
participants
and personnel
(performance
bias)

Blinding of           (b)             (c)             (b)
outcome
assessment
(detection
bias)

Incomplete            (a)             (a)             (a)
outcome data
(attrition
bias)

Selective             (b)             (b)             (b)
reporting
(reporting
bias)

Other bias            (c)             (a)             (a)

                  Du SG 2017   Chen Z 2013   Chen XZ 2016

Random sequence      (a)           (b)           (b)
generation
(selection
bias)

Allocation           (b)           (b)           (b)
concealment
(selection
bias)

Blinding of          (b)           (c)           (b)
participants
and personnel
(performance
bias)

Blinding of          (b)           (c)           (b)
outcome
assessment
(detection
bias)

Incomplete           (a)           (a)           (a)
outcome data
(attrition
bias)

Selective            (b)           (b)           (b)
reporting
(reporting
bias)

Other bias           (c)           (a)           (a)

Legend:

(a) Low risk of bias

(b) Unclear risk of bias

(c) High risk of bias

Figure 4: TEs between two groups.

                            Experimental          Control
Study or subgroup
                          Events     Total    Events     Total

Chen XZ, 2016               37        40        28        40
Chen Z, 2013                99        105       78        89
Du SG, 2017                 51        55        42        51
Huang CB, 2013              34        40        30        40
Huang RY, 2019              64        72        38        46
Jiang P, 2016               26        31        17        28
Li EK, 2007                  4        28         3        30
Lin CS, 2011                70        79        54        79
Liu W, 2007                 55        60        48        60
Lu SJ, 2002                 39        40        12        20
Qian X, 2015                77        84        59        84
Wang AY, 2016               27        28        20        28
Wang JM, 2013               96        120       80        120
Wang Z, 2014                43        47        31        41
Wu YJ, 2001                 35        35        35        35
Yang M, 2017                69        69        58        58
Yu SY, 2013                 113       120       49        60
Zhang YY, 2016              30        35        21        33
Zhao SS, 2012               38        64        14        40
Zhao YX, 2006               39        40        34        40

Total (95% CI)                       1192                1022
Total events               1046                 751

Study or subgroup         Weight         Odds ratio
                            (%)      M-H, fixed, 95% CI

Chen XZ, 2016               2.8      5.29 [1.36, 20.53]
Chen Z, 2013                6.4      2.33 [0.82, 6.57]
Du SG, 2017                 4.2      2.73 [0.79, 9.50]
Huang CB, 2013              6.0      1.89 [0.61, 5.82]
Huang RY, 2019              6.9      1.68 [0.58, 4.86]
Jiang P, 2016               3.8      3.36 [0.99, 11.41]
Li EK, 2007                 3.3      1.50 [0.30, 7.39]
Lin CS, 2011                8.2      3.60 [1.55, 8.35]
Liu W, 2007                 5.3      2.75 [0.90, 8.37]
Lu SJ, 2002                 0.5     26.00 [2.95, 229.36]
Qian X, 2015                6.6      4.66 [1.89, 11.51]
Wang AY, 2016               1.0     10.80 [1.25, 93.44]
Wang JM, 2013              21.4      2.00 [1.11, 3.60]
Wang Z, 2014                3.8      3.47 [1.00, 12.08]
Wu YJ, 2001                            Not estimable
Yang M, 2017                           Not estimable
Yu SY, 2013                 5.1      3.62 [1.33, 9.90]
Zhang YY, 2016              4.1      3.43 [1.05, 11.19]
Zhao SS, 2012               9.4      2.71 [1.20, 6.16]
Zhao YX, 2006               1.1      6.88 [0.79, 60.06]

Total (95% CI)             100.0     3.03 [2.36, 3.88]
Total events

Heterogeneity: [chi square] = 12.45, df = 17 (P = 0.77);
[I.sup.2] = 0%

Test for overall effect Z = 8.76 (P < 0.00001)

Figure 5: TJC between two groups.

                                 Experimental
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013              2.2       1.2       40
Huang RY, 2019             2.89      3.49       72
Lin CS, 2011               1.41      1.64       79
Liu W, 2007                8.06      3.18       60
Lu SJ, 2002                 1.7       1.2       40
Wang JM, 2013              3.62      1.46       120
Wang Z, 2014               3.17      2.43       47
Wu YJ, 2001                 6.5        1        35
Yang M, 2017               1.08      1.83       69
Yu SY, 2013                7.23      7.25       120

Total (95% CI)                                  682

                                     Control
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013             3.87      1.68       40
Huang RY, 2019             2.43      1.94       46
Lin CS, 2011               3.06      2.17       79
Liu W, 2007                8.25      3.64       60
Lu SJ, 2002                 4.7      2.54       20
Wang JM, 2013              4.57      4.21       120
Wang Z, 2014               5.97      3.73       41
Wu YJ, 2001                 5.2       1.1       35
Yang M, 2017               1.79      2.16       58
Yu SY, 2013                10.74     6.88       60

Total (95% CI)                                  559

Study or subgroup         Weight       Mean difference
                            (%)      IV, random, 95% CI

Huang CB, 2013             10.8     -1.67 [-2.31, -1.03]
Huang RY, 2019             10.9      0.46 [-0.52, 1.44]
Lin CS, 2011               10.2     -1.65 [-2.25, -1.05]
Liu W, 2007                 9.6      -0.19 [-1.41, 1.03]
Lu SJ, 2002                 9.7     -3.00 [-4.17, -1.83]
Wang JM, 2013              10.5     -0.95 [-1.75, -0.15]
Wang Z, 2014                9.3     -2.80 [-4.14, -1.46]
Wu YJ, 2001                11.1       1.30 [0.81, 1.79]
Yang M, 2017               10.7     -0.71 [-1.41, -0.01]
Yu SY, 2013                 7.2     -3.51 [-5.68, -1.34]

Total (95% CI)             100.0    -1.17 [-2.12, -0.21]

Heterogeneity: [tau.sup.2] = 2.09; [chi square] = 118.69,
df = 9 (P < 0.00001); [I.sup.2] = 92%

Test for overall effect: Z = 2.40 (P = 0.02)

Figure 6: SJC between two groups.

                                 Experimental
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013              2.1       1.2       40
Huang RY, 2019             0.44      1.16       72
Lin CS, 2011               0.85      1.09       79
Liu W, 2007                7.68      3.25       60
Lu SJ, 2002                1.48      1.15       40
Wang JM, 2013              2.95      1.38       120
Wang Z, 2014               2.17      1.97       47
Wu YJ, 2001                 5.5       0.9       35
Yang M, 2017               0.79      0.99       69
Yu SY, 2013                4.26      3.22       120

Total (95% CI)                                  682

                                    Control
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013              4.7       1.8       40
Huang RY, 2019             0.46      0.84       46
Lin CS, 2011               2.11      2.62       79
Liu W, 2007                7.49      3.39       60
Lu SJ, 2002                2.65      1.66       20
Wang JM, 2013              3.52      2.48       120
Wang Z, 2014               4.57      3.93       41
Wu YJ, 2001                 3.5       0.6       35
Yang M, 2017               1.46      1.77       58
Yu SY, 2013                 7.1      5.88       60

Total (95% CI)                                  559

Study or subgroup         Weight      Mean difference
                            (%)      IV, random, 95% CI

Huang CB, 2013             10.3     -2.60 [-3.27, -1.93]
Huang RY, 2019             10.7     -0.02 [-0.38, 0.34]
Lin CS, 2011               10.4     -1.26 [-1.89, -0.63]
Liu W, 2007                 9.3      0.19 [-1.00, 1.38]
Lu SJ, 2002                10.1     -1.17 [-1.98, -0.36]
Wang JM, 2013              10.5     -0.57 [-1.08, -0.06]
Wang Z, 2014                9.0     -2.40 [-3.73, -1.07]
Wu YJ, 2001                10.7      2.00 [1.64, 2.36]
Yang M, 2017               10.5     -0.67 [-1.18, -0.16]
Yu SY, 2013                 8.4     -2.84 [-4.44, -1.24]

Total (95% CI)             100.0    -0.87 [-1.85, 0.10]

Heterogeneity: [tau.sup.2] = 2.29; [chi square] = 236.51, df = 9
(P < 0.00001); [I.sup.2] = 96%

Test for overall effect: Z = 1.75 (P = 0.08)

Figure 7: DMS between two groups.

                                 Experimental
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013              1.3       1.8       40
Huang RY, 2019             0.94      0.84       72
Lin CS, 2011               6.42      10.14      79
Lu SJ, 2002                12.75      21        40
Wang JM, 2013              19.82     16.54      120
Wang Z, 2014               15.37     18.59      47
Wu YJ, 2001                 24         7        35
Yu SY, 2013                1.25      0.83       120
Zhao YX, 2006                1        0.5       40

Total (95% CI)                                  593

                                    Control
Study or subgroup
                           Mean       SD       Total

Huang CB, 2013              3.7       2.4       40
Huang RY, 2019             1.41      1.68       46
Lin CS, 2011               12.76     16.95      79
Lu SJ, 2002                33.5      44.52      20
Wang JM, 2013              21.23     19.37      120
Wang Z, 2014               42.87     30.89      41
Wu YJ, 2001                 30        10        35
Yu SY, 2013                1.76      0.87       60
Zhao YX, 2006               1.8       0.6       40

Total (95% CI)                                  481

Study or subgroup         Weight    Std. mean difference
                            (%)      IV, random, 95% CI

Huang CB, 2013             10.4     -1.12 [-1.59, -0.65]
Huang RY, 2019             11.6     -0.38 [-0.75, -0.00]
Lin CS, 2011               12.3     -0.45 [-0.77, -0.14]
Lu SJ, 2002                 9.4     -0.67 [-1.22, -0.12]
Wang JM, 2013              13.0      -0.08 [-0.33, 0.18]
Wang Z, 2014               10.6     -1.09 [-1.54, -0.64]
Wu YJ, 2001                10.2     -0.69 [-1.17, -0.20]
Yu SY, 2013                12.3     -0.60 [-0.92, -0.29]
Zhao YX, 2006              10.1     -1.43 [-1.93, -0.94]

Total (95% CI)             100.0    -0.69 [-0.98, -0.41]

Heterogeneity: [tau.sup.2] = 0.14; [chi square] = 38.48, df = 8
(P < 0.00001); [I.sup.2] = 79%

Test for overall effect: Z = 4.78 (P < 0.00001)

Figure 8: GS between two groups.

                                 Experimental
Study or subgroup
                           Mean       SD       Total

Lu SJ, 2002               142.63     26.51      40
Wang JM, 2013              79.03     22.48      120
Wu YJ, 2001                 150        9        35
Yu SY, 2013                9.04      2.67       120
Zhao YX, 2006             203.92     67.7       40

Total (95% CI)                                  355

                                    Control
Study or subgroup
                           Mean       SD       Total

Lu SJ, 2002                117.5     28.26      20
Wang JM, 2013              77.8      16.92      120
Wu YJ, 2001                 170       12        35
Yu SY, 2013                6.86      2.59       60
Zhao YX, 2006             165.21     54.19      40

Total (95% CI)                                  275

Study or subgroup         Weight    Std. mean difference
                            (%)      IV, random, 95% CI

Lu SJ, 2002                19.2      0.92 [0.35, 1.48]
Wang JM, 2013              21.0      0.06 [-0.19, 0.31]
Wu YJ, 2001                19.2     -1.86 [-2.43, -1.30]
Yu SY, 2013                20.7      0.61 [0.29, 0.92]
Zhao YX, 2006              19.9      0.82 [0.36, 1.28]

Total (95% CI)             100.0     0.12 [-0.63, 0.87]

Heterogeneity: [tau.sup.2] = 0.68; [chi square] = 71.81, df = 4
(P < 0.00001); [I.sup.2] = 94%

Test for overall effect: Z = 0.32 (P = 0.75)

Figure 9: DAS28 between two groups.

                                 Experimental
Study or subgroup
                           Mean       SD       Total

Du SG, 2017                2.48      0.87       55
Huang CB, 2013              4.1       0.7       40
Huang RY, 2019             2.58      1.14       72
Jiang P, 2016               2.6       0.4       31
Wang JM, 2013              2.96      1.09       120

Total (95% CI)                                  318

                                    Control
Study or subgroup
                           Mean       SD       Total

Du SG, 2017                3.17      1.16       51
Huang CB, 2013              4.8       0.7       40
Huang RY, 2019             2.74      0.97       46
Jiang P, 2016               2.9       0.5       28
Wang JM, 2013              3.37      1.46       120

Total (95% CI)                                  285

Study or subgroup         Weight       Mean difference
                            (%)       IV, fixed, 95% CI

Du SG, 2017                12.5     -0.69 [-1.08, -0.30]
Huang CB, 2013             20.5     -0.70 [-1.01, -0.39]
Huang RY, 2019             13.1      -0.16 [-0.54, 0.22]
Jiang P, 2016              35.7     -0.30 [-0.53, -0.07]
Wang JM, 2013              18.2     -0.41 [-0.74, -0.08]

Total (95% CI)             100.0    -0.43 [-0.57, -0.29]

Heterogeneity: [chi square] = 7.77, df = 4 (P = 0.10);
[I.sup.2] = 48%

Test for overall effect: Z = 6.10 (P < 0.00001)

Figure 10: RF between two groups.

                                 Experimental

Study or subgroup          Mean       SD       Total

Du SG, 2017                32.03     21.24      55
Huang CB, 2013             89.3      31.5       40
Jiang P, 2016               1.9       0.4       31
Lin CS, 2011               79.35     93.74      79
Lu SJ, 2002                80.39     6.55       40
Wang Z, 2014               66.87     72.48      47
Wu YJ, 2001                 66        23        35
Yang M, 2017               41.65     32.38      69
Yu SY, 2013                70.24     46.25      120
Zhang YY, 2016             1.89      0.39       35
Zhao SS, 2012              42.6      17.8       64
Zhao YX, 2006              38.56     14.35      40

Total (95% CI)                                  655

                                    Control

Study or subgroup          Mean       SD       Total

Du SG, 2017                43.36     30.32      51
Huang CB, 2013             112.2      40        40
Jiang P, 2016               1.9       0.4       28
Lin CS, 2011              109.22    132.65      79
Lu SJ, 2002               118.73     10.72      20
Wang Z, 2014              124.65    134.78      41
Wu YJ, 2001                 82        28        35
Yang M, 2017               74.31     96.61      58
Yu SY, 2013                87.78     56.11      60
Zhang YY, 2016             1.92      0.37       33
Zhao SS, 2012              46.1      15.2       40
Zhao YX, 2006              46.37     15.34      40

Total (95% CI)                                  525

Study or subgroup         Weight    Std. mean difference
                            (%)      IV, random, 95% CI

Du SG, 2017                 8.8     -0.43 [-0.82, -0.05]
Huang CB, 2013              8.4     -0.63 [-1.08, -0.18]
Jiang P, 2016               8.0      0.00 [-0.51, 0.51]
Lin CS, 2011                9.2      -0.26 [-0.57, 0.05]
Lu SJ, 2002                 5.0     -4.64 [-5.65, -3.63]
Wang Z, 2014                8.6     -0.54 [-0.97, -0.11]
Wu YJ, 2001                 8.2     -0.62 [-1.10, -0.14]
Yang M, 2017                9.0     -0.47 [-0.82, -0.11]
Yu SY, 2013                 9.2     -0.35 [-0.66, -0.04]
Zhang YY, 2016              8.3      -0.08 [-0.55, 0.40]
Zhao SS, 2012               8.8      -0.21 [-0.60, 0.19]
Zhao YX, 2006               8.4     -0.52 [-0.97, -0.07]

Total (95% CI)             100.0    -0.59 [-0.91, -0.27]

Heterogeneity: [tau.sup.2] = 0.26; [chi square] = 75.63, df = 11
(P < 0.00001); [I.sup.2] = 85%

Test for overall effect: Z = 3.60 (P = 0.0003)

Figure 11: Anti-CCP between two groups.

                                 Experimental

Study or subgroup          Mean       SD       Total

Du SG, 2017                1.84      5.38       55
Jiang P, 2016               2.5       0.4       31
Lin CS, 2011               40.06     49.29      79
Yang M, 2017               44.59     24.45      69
Zhang YY, 2016             2.54      0.35       35
Zhao SS, 2012              84.2      44.6       64
Zhao YX, 2006              46.37     20.13      40

Total (95% CI)                                  373

                                    Control

Study or subgroup          Mean       SD       Total

Du SG, 2017                2.18      8.43       51
Jiang P, 2016               2.6       0.4       28
Lin CS, 2011               51.28     69.52      79
Yang M, 2017               63.41     55.54      58
Zhang YY, 2016              2.6      0.38       33
Zhao SS, 2012              85.6      40.3       40
Zhao YX, 2006              54.34     28.67      40

Total (95% CI)                                  329

                          Weight    Std. mean difference
Study or subgroup           (%)     IV, fixed, 95% CI

Du SG, 2017                15.3     -0.05 [-0.43, 0.33]
Jiang P, 2016               8.5     -0.25 [-0.76, 0.27]
Lin CS, 2011               22.8     -0.19 [-0.50, 0.13]
Yang M, 2017               17.8     -0.45 [-0.80, -0.10]
Zhang YY, 2016              9.8     -0.16 [-0.64, 0.31]
Zhao SS, 2012              14.3     -0.03 [-0.43, 0.36]
Zhao YX, 2006              11.5     -0.32 [-0.76, 0.12]

Total (95% CI)             100.0    -0.21 [-0.36, -0.06]

Heterogeneity: [chi square] = 3.55, df = 6 (P = 0.74);
[I.sup.2] = 0%

Test for overall effect: Z = 2.73 (P = 0.006)

Figure 12: ESR between two groups.

                                 Experimental

Study or subgroup          Mean       SD       Total

Du SG, 2017                19.73     12.27      55
Huang CB, 2013              26        9.2       40
Huang RY, 2019             31.54     21.08      72
Jiang P, 2016               27        11        31
Lin CS, 2011               21.52     10.54      79
Liu W, 2007                29.14     15.19      60
Lu SJ, 2002                32.48     21.45      40
Wang JM, 2013              18.62     9.78       120
Wang Z, 2014               13.38      8.7       47
Wu YJ, 2001                 40         5        35
Yang M, 2017               21.03     12.57      69
Yu SY, 2013                34.26     21.12      120
Zhang YY, 2016             27.26     11.45      35
Zhao SS, 2012              29.5      13.4       64
Zhao YX, 2006              25.64     6.41       40

Total (95% CI)                                  907

                                    Control

Study or subgroup          Mean       SD       Total

Du SG, 2017                29.34     14.35      51
Huang CB, 2013             58.6      19.2       40
Huang RY, 2019             41.04     28.86      46
Jiang P, 2016               38        21        28
Lin CS, 2011               27.44     16.83      79
Liu W, 2007                34.08     17.25      60
Lu SJ, 2002                45.95     22.26      20
Wang JM, 2013              21.21     15.26      120
Wang Z, 2014               19.53     17.03      41
Wu YJ, 2001                 36         3        35
Yang M, 2017               25.16     15.58      58
Yu SY, 2013                45.15     23.23      60
Zhang YY, 2016             38.45     20.5       33
Zhao SS, 2012              40.5      20.8       40
Zhao YX, 2006              28.48     7.58       40

Total (95% CI)                                  751

Study or subgroup         Weight        Mean difference
                            (%)       IV, random, 95% CI

Du SG, 2017                 7.1      -9.61 [-14.71, -4.51]
Huang CB, 2013              6.6     -32.60 [-39.20, -26.00]
Huang RY, 2019              5.6      -9.50 [-19.16, 0.16]
Jiang P, 2016               5.9     -11.00 [-19.69, -2.31]
Lin CS, 2011                7.3      -5.92 [-10.30, -1.54]
Liu W, 2007                 6.9      -4.94 [-10.76, 0.88]
Lu SJ, 2002                 4.9     -13.47 [-25.28, -1.66]
Wang JM, 2013               7.5       -2.59 [-5.83, 0.65]
Wang Z, 2014                6.9      -6.15 [-11.93, -0.37]
Wu YJ, 2001                 7.7        4.00 [2.07, 5.93]
Yang M, 2017                7.1       -4.13 [-9.12, 0.86]
Yu SY, 2013                 6.5     -10.89 [-17.88, -3.90]
Zhang YY, 2016              6.2     -11.19 [-19.15, -3.23]
Zhao SS, 2012               6.4     -11.00 [-18.23, -3.77]
Zhao YX, 2006               7.6       -2.84 [-5.92, 0.24]

Total (95% CI)             100.0     -8.36 [-12.60, -4.12]

Heterogeneity: [tau.sup.2] = 59.48; [chi square] = 159.43, df = 14
(P < 0.00001); [I.sup.2] = 91%

Test for overall effect: Z = 3.86 (P = 0.0001)

Figure 13: CRP between two groups.

                                 Experimental

Study or subgroup          Mean       SD       Total

Du SG, 2017                9.19      5.72       55
Huang CB, 2013              7.6       3.9       40
Huang RY, 2019              7.3      10.31      72
Jiang P, 2016               10         6        31
Lin CS, 2011               6.01      9.12       79
Liu W, 2007                1.45      0.88       60
Lu SJ, 2002                  7       1.25       40
Wang JM, 2013              8.71      5.39       120
Wang Z, 2014               3.16      6.93       47
Yang M, 2017               25.3      20.2       69
Yu SY, 2013                35.53     30.15      120
Zhang YY, 2016             9.79      6.18       35
Zhao SS, 2012               7.5       3.4       64
Zhao YX, 2006              23.34     6.31       40

Total (95% CI)                                  872

                                    Control

Study or subgroup          Mean       SD       Total

Du SG, 2017                17.65     7.36       51
Huang CB, 2013             27.8        6        40
Huang RY, 2019             11.84     17.17      46
Jiang P, 2016               16         7        28
Lin CS, 2011               7.51      6.62       79
Liu W, 2007                 1.5       0.8       60
Lu SJ, 2002                9.18      1.38       20
Wang JM, 2013              9.56       8.6       120
Wang Z, 2014               8.64      12.24      41
Yang M, 2017               37.8      27.73      58
Yu SY, 2013                57.4      30.62      60
Zhang YY, 2016             15.85      7.6       33
Zhao SS, 2012              18.5       7.5       40
Zhao YX, 2006              26.67     6.92       40

Total (95% CI)                                  716

Study or subgroup         Weight        Mean difference
                            (%)       IV, random, 95% CI

Du SG, 2017                 7.8      -8.46 [-10.98, -5.94]
Huang CB, 2013              7.9     -20.20 [-22.42, -17.98]
Huang RY, 2019              6.2      -4.54 [-10.04, 0.96]
Jiang P, 2016               7.4      -6.00 [-9.34, -2.66]
Lin CS, 2011                7.8       -1.50 [-3.99, 0.99]
Liu W, 2007                 8.4       -0.05 [-0.35, 0.25]
Lu SJ, 2002                 8.3      -2.18 [-2.90, -1.46]
Wang JM, 2013               8.1       -0.85 [-2.67, 0.97]
Wang Z, 2014                6.9      -5.48 [-9.72, -1.24]
Yang M, 2017                4.5     -12.50 [-21.08, -3.92]
Yu SY, 2013                 4.1     -21.87 [-31.31, -12.43]
Zhang YY, 2016              7.4      -6.06 [-9.36, -2.76]
Zhao SS, 2012               7.8     -11.00 [-13.47, -8.53]
Zhao YX, 2006               7.6      -3.33 [-6.23, -0.43]

Total (95% CI)             100.0     -6.73 [-9.38, -4.08]

Heterogeneity: [tau.sup.2] = 21.83; [chi square] = 484.94, df = 13
(P < 0.00001); [I.sup.2] = 97%

Test for overall effect: Z = 4.98 (P < 0.00001)

Figure 14: AEs between two groups.

                            Experimental           Control
Study or subgroup
                          Events     Total    Events     Total

11.1.1 Gastrointestinal disorders
  Du SG, 2017                4        56         7        56
  Huang CB, 2013             0        40         4        40
  Huang RY, 2019             7        73        12        47
  Jiang P, 2016              1        32         2        31
  Li EK, 2007                4        32         5        33
  Lin CS, 2011               5        79         6        79
  Liu W, 2007                0        60        13        60
  Lu SJ, 2002                4        40         6        20
  Qian X, 2015               3        84         5        84
  Wang AY, 2016              1        28         1        28
  Wang JM, 2013              6        120       13        120
  Wu YJ, 2001                5        35        12        35
  Yang M, 2017               1        79         5        80
  Zhang YY, 2016             0        36         1        36
  Zhao SS, 2012              4        64         5        40
  Zhao YX, 2006              6        40         5        40
Subtotal (95% CI)                     898                 829
Study or subgroup
Total events                51                  102

Heterogeneity: [chi square] = 11.41, df = 15 (P = 0.72);
[I.sup.2] = 0%

Test for overall effect: Z = 5.03 (P < 0.00001)

11.1.2 Abnormal liver function
  Chen Z, 2013              10        105       31        89
  Du SG, 2017                1        56         3        56
  Huang CB, 2013             0        40         2        40
  Huang RY, 2019             4        73         9        47
  Jiang P, 2016              1        32         2        31
  Lin CS, 2011               3        79         4        79
  Liu W, 2007                0        60         4        60
  Lu SJ, 2002                0        40         2        20
  Qian X, 2015               2        84         4        84
  Wang AY, 2016              0        120        5        120
  Wang Z, 2014               0        47         4        41
  Wu YJ, 2001                2        35         3        35
  Yang M, 2017               0        79         3        80
  Zhang YY, 2016             1        36         2        36
  Zhao SS, 2012              2        64         4        40
Subtotal (95% CI)                     950                 858
Total events                26                  82

Heterogeneity: [chi square] = 6.38, df = 14 (P = 0.96);
[I.sup.2] = 0%

Test for overall effect: Z = 6.38 (P < 0.00001)

11.1.3 Leukopenia
  Du SG, 2017                1        56         2        56
  Huang RY, 2019             1        73         1        47
  Liu W, 2007                0        60         1        60
  Lu SJ, 2002                0        40         1        20
  Qian X, 2015               1        84         1        84
  Wang JM, 2013              1        120        1        120
  Wang Z, 2014               0        47         5        41
  Wu YJ, 2001                0        35         2        35
  Zhang YY, 2016             0        36         3        36
  Zhao SS, 2012              0        64         2        40
  Zhao YX, 2006              0        40         2        40
Subtotal (95% CI)                     655                 579
Total events                 4                  21

Heterogeneity: [chi square] = 1.87, df = 10
(P = 1.00); [I.sup.2] = 0%

Test for overall effect: Z = 5.03 (P = 0.001)

11.1.4 Skin allergies and rashes
  Du SG, 2017                0        56         1        56
  Huang RY, 2019             8        32         1        47
  Jiang P, 2016              0        32         1        31
  Li EK, 2007                0        120        1        33
  Wang JM, 2013              0        47         1        120
  Wang Z, 2014               0        36         2        41
  Zhang YY, 2016             1        40         2        36
  Zhao YX, 2006              0        436        2        40
Subtotal (95% CI)                     73                  404
Total events                 9                  11

Heterogeneity: [chi square] = 6.26, df = 7 (P = 0.51);
[I.sup.2] = 0%

Test for overall effect: Z = 0.85 (P = 0.40)

11.1.5 Headaches and dizziness
  Li EK, 2007                1        32         0        33
  Lin CS, 2011               2        79         0        79
  Wang JM, 2013              0        120        1        120
  Zhao YX, 2006              2        40         3        40
Subtotal (95% CI)                     271                 272
Total events                 5                   2

Heterogeneity: [chi square] = 1.83, df = 3 (P = 0.61);
[I.sup.2] = 0%

Test for overall effect: Z = 0.62 (P = 0.54)

11.1.6 Alopecia
  Huang RY, 2019             1        73         4        47
  Lin CS, 2011               2        79         3        79
  Wang JM, 2013              0        120        1        120
  Wu YJ, 2001                0        35         9        35
Subtotal (95% CI)                     307                 281
Total events                 3                   1

Heterogeneity: [chi square] = 1.04, df = 3 (P = 0.79);
[I.sup.2] = 0%

Test for overall effect: Z = 1.83 (P = 0.07)

Total (95% CI)                       3517                3223
Total events                98                  228

Heterogeneity: [chi square] = 36.12, df = 57 (P = 0.99);
[I.sup.2] = 0%

Test for overall effect: Z = 8.32 (P < 0.00001)

Test for subgroup differences: [chi square] = 11.22, df = 5
(P = 0.05) [I.sup.2] = 55.4%

Study or subgroup                   Weight        Odds ratio
                                      (%)     M-H, fixed, 95% CI

11.1.1 Gastrointestinal disorders
  Du SG, 2017                         2.7      0.54 [0.15, 1.95]
  Huang CB, 2013                      1.8      0.10 [0.01, 1.92]
  Huang RY, 2019                      5.4      0.31 [0.11, 0.86]
  Jiang P, 2016                       0.8      0.47 [0.04, 5.44]
  Li EK, 2007                         1.8      0.80 [0.19, 3.29]
  Lin CS, 2011                        2.3      0.82 [0.24, 2.81]
  Liu W, 2007                         5.5      0.03 [0.00, 0.50]
  Lu SJ, 2002                         3.0      0.26 [0.06, 1.06]
  Qian X, 2015                        2.0      0.59 [0.14, 2.53]
  Wang AY, 2016                       0.4     1.00 [0.06, 16.82]
  Wang JM, 2013                       5.1      0.43 [0.16, 1.18]
  Wu YJ, 2001                         4.2      0.32 [0.10, 1.04]
  Yang M, 2017                        2.0      0.19 [0.02, 1.68]
  Zhang YY, 2016                      0.6      0.32 [0.01, 8.23]
  Zhao SS, 2012                       2.4      0.47 [0.12, 1.85]
  Zhao YX, 2006                       1.8      1.24 [0.34, 4.43]
Subtotal (95% CI)                    41.8      0.40 [0.28, 0.58]
Study or subgroup
Total events

Heterogeneity: [chi square] = 11.41, df = 15 (P = 0.72);
[I.sup.2] = 0%

Test for overall effect: Z = 5.03 (P < 0.00001)

11.1.2 Abnormal liver function
  Chen Z, 2013                       12.5      0.20 [0.09, 0.43]
  Du SG, 2017                         1.2      0.32 [0.03, 3.19]
  Huang CB, 2013                      1.0      0.19 [0.01, 4.09]
  Huang RY, 2019                      4.3      0.24 [0.07, 0.85]
  Jiang P, 2016                       0.8      0.47 [0.04, 5.44]
  Lin CS, 2011                        1.6      0.74 [0.16, 3.42]
  Liu W, 2007                         1.8      0.10 [0.01, 1.97]
  Lu SJ, 2002                         1.3      0.09 [0.00, 2.00]
  Qian X, 2015                        1.6      0.49 [0.09, 2.74]
  Wang AY, 2016                       2.3      0.09 [0.00, 1.59]
  Wang Z, 2014                        2.0      0.09 [0.00, 1.68]
  Wu YJ, 2001                         1.2      0.65 [0.10, 4.13]
  Yang M, 2017                        1.4      0.14 [0.01, 2.74]
  Zhang YY, 2016                      0.8      0.49 [0.04, 5.61]
  Zhao SS, 2012                       2.0      0.29 [0.05, 1.66]
Subtotal (95% CI)                    35.7      0.25 [0.16, 0.39]
Total events

Heterogeneity: [chi square] = 6.38, df = 14 (P = 0.96);
[I.sup.2] = 0%

Test for overall effect: Z = 6.38 (P < 0.00001)

11.1.3 Leukopenia
  Du SG, 2017                         0.8      0.49 [0.04, 5.57]
  Huang RY, 2019                      0.5     0.64 [0.04, 10.47]
  Liu W, 2007                         2.2      0.08 [0.00, 1.54]
  Lu SJ, 2002                         0.8      0.16 [0.01, 4.12]
  Qian X, 2015                        0.8      0.49 [0.04, 5.55]
  Wang JM, 2013                       1.4      0.14 [0.01, 2.73]
  Wang Z, 2014                        0.7      0.28 [0.01, 7.17]
  Wu YJ, 2001                         0.6      0.32 [0.01, 8.23]
  Zhang YY, 2016                      0.6      0.32 [0.01, 8.23]
  Zhao SS, 2012                       1.0      0.30 [0.03, 3.44]
  Zhao YX, 2006                       1.0      0.19 [0.01, 4.09]
Subtotal (95% CI)                    10.5      0.26 [0.11, 0.59]
Total events

Heterogeneity: [chi square] = 1.87, df = 10
(P = 1.00); [I.sup.2] = 0%

Test for overall effect: Z = 5.03 (P = 0.001)

11.1.4 Skin allergies and rashes
  Du SG, 2017                         0.6      0.33 [0.01, 8.21]
  Huang RY, 2019                      0.4     5.66 [0.68, 46.83]
  Jiang P, 2016                       0.6      0.31 [0.01, 7.98]
  Li EK, 2007                         0.6      0.33 [0.01, 8.49]
  Wang JM, 2013                       0.6      0.33 [0.01, 8.20]
  Wang Z, 2014                        1.1      0.17 [0.01, 3.57]
  Zhang YY, 2016                      0.8      0.49 [0.04, 5.61]
  Zhao YX, 2006                       1.0      0.19 [0.01, 4.09]
Subtotal (95% CI)                     5.8      0.71 [0.31, 1.58]
Total events

Heterogeneity: [chi square] = 6.26, df = 7 (P = 0.51);
[I.sup.2] = 0%

Test for overall effect: Z = 0.85 (P = 0.40)

11.1.5 Headaches and dizziness
  Li EK, 2007                         0.2     3.19 [0.13, 81.25]
  Lin CS, 2011                        0.2     5.13 [0.24, 108.57]
  Wang JM, 2013                       0.6      0.33 [0.01, 8.20]
  Zhao YX, 2006                       0.8      1.00 [0.13, 7.47]
Subtotal (95% CI)                     1.8      1.47 [0.43, 4.96]
Total events

Heterogeneity: [chi square] = 1.83, df = 3 (P = 0.61);
[I.sup.2] = 0%

Test for overall effect: Z = 0.62 (P = 0.54)

11.1.6 Alopecia
  Huang RY, 2019                      2.0      0.15 [0.02, 1.38]
  Lin CS, 2011                        1.2      0.66 [0.11, 4.05]
  Wang JM, 2013                       0.6      0.33 [0.01, 8.20]
  Wu YJ, 2001                         0.6      0.32 [0.01, 8.23]
Subtotal (95% CI)                     4.4      0.34 [0.11, 1.08]
Total events

Heterogeneity: [chi square] = 1.04, df = 3 (P = 0.79);
[I.sup.2] = 0%

Test for overall effect: Z = 1.83 (P = 0.07)

Total (95% CI)                       100.0     0.37 [0.29, 0.47]
Total events

Heterogeneity: [chi square] = 36.12, df = 57 (P = 0.99);
[I.sup.2] = 0%

Test for overall effect: Z = 8.32 (P < 0.00001)

Test for subgroup differences: [chi square] = 11.22, df = 5
(P = 0.05) [I.sup.2] = 55.4%
COPYRIGHT 2020 Hindawi Limited
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2020 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Xing, Qi; Fu, Ling; Yu, Zhichao; Zhou, Xueping
Publication:Evidence - Based Complementary and Alternative Medicine
Geographic Code:9CHIN
Date:Apr 30, 2020
Words:13780
Previous Article:Electroacupuncture Ameliorates Acute Myocardial Ischemia: A Potential Role of the Locus Coeruleus.
Next Article:Involvement of Cathepsins in Innate and Adaptive Immune Responses in Periodontitis.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |