Combined traditional Chinese medicine therapy for the treatment of infertility with polycystic ovary syndrome: A network meta-analysis of randomized controlled trials

Abstract

Background:

Polycystic ovary syndrome (PCOS) infertility has attracted great attention from researchers due to its high incidence. Numerous studies have shown that Chinese medicine is effective in treating this disease, but there is a wide variety of Chinese medicine therapies available, and there is a lack of comparative evaluation of the efficacy of various Chinese medicine combination therapies in the clinic, which requires further in-depth exploration. This study aims to evaluate the efficacy of a combined traditional Chinese medicine (TCM) therapy for the treatment of infertility with PCOS using network meta-analysis (NMA).

Methods:

In PubMed, web of Science, Cochrane Library, Embase, China Knowledge Network, Wanfang Data, VIP Database, China Biomedical Literature Database (SinoMed) databases, searchs were conducted for information about the randomized controlled trials (RCTs) of combined TCM therapy for the treatment of infertility with PCOS. Quality evaluation was performed using the Cochrane 5.3 risk of bias assessment tool, and NMA using Stata 16.0.

Results:

This study comprised 28 RCTs using 8 combined TCM therapies in total. The results of the NMA showed that moxibustion + herbal, fire acupuncture + herbal, acupuncture + herbal, electroacupuncture + herbal, and acupoint application + herbal improved the clinical pregnancy rate better than acupuncture, herbal, and western medicines monotherapy (P < .05). Additionally, ear point pressure + herbal enema + herbal, acupuncture and moxibustion + herbal, fire acupuncture + herbal, and acupuncture + herbal improved the ovulation rate better than acupuncture, herbal, and western medicines monotherapy (P < .05). Moxibustion + herbal, fire acupuncture + herbal, and acupuncture + herbal are the 3 most effective therapies for improving the clinical pregnancy rate. Fire acupuncture + herbal, acupuncture + herbal, and ear point pressure + herbal enema + herbal are the 3 most effective therapies for improving the ovulation rate.

Conclusion:

The combined TCM therapy demonstrated better efficacy for the treatment of infertility with PCOS compared to acupuncture, herbal, and western medicines monotherapy. However, the optimal treatment therapy varied depending on the outcome indicators. Further large sample, high-quality, and standardized RCTs are needed to verify these findings.

1. Introduction

Polycystic ovary syndrome (PCOS) is a clinical condition characterized by hyperandrogenism, oligoovulation, and polycystic ovarian morphology.^[1] It is among the major challenges of modern society, with a complex etiology, and is a common gynecological disorder that significantly impacts the reproductive health and quality of life of women of reproductive age. The estimated prevalence of PCOS before menopause is 1 in 10 women, ranging from 5% to 15% depending on ethnicity, and affects 7% to 15% of women of reproductive age. In China, the prevalence rate is about 5.6% among women aged 19 to 45 years.^[2–5] Infertility due to PCOS accounts for approximately 30% of all female infertility cases and is the leading cause of anovulatory infertility.^[6,7] The World Health Organization predicts that infertility will be among the top 3 common diseases in the 21st century following tumors, cardiovascular, and cerebrovascular diseases, with a global prevalence rate of 8% to 12%.^[8,9] It has been reported^[10] that in China, the number of infertility patients has exceeded 50 million accounting for about 15% of women of reproductive age. The decline in female fertility due to aging will present numerous challenges to the country’s economy and society. Therefore, infertility caused by PCOS has become a significant research topic.

For the management of infertility PCOS in clinical, western drugs typically employ drugs such as letrozole, clomiphene citrate, and clomiphene. However, these drugs carry risks such as ovarian overstimulation, adverse reactions such as hot flashes, nausea, and breast pain, as well as drug resistance.^[11–13] Alternative, non-pharmacological treatment options such as ultrasound-guided follicular puncture and laparoscopic ovarian drilling also entail certain risks and may lead to postoperative complications like periovarian adhesions and ovarian hypoplasia.^[14] While assisted reproductive technologies are already in clinical use, such as intrauterine insemination and in vitro maturation culture of immature eggs are available, their efficacy remains a matter of debate; and often lead to high ovulation rate, low pregnancy rate, high miscarriage rate.^[11,15] In addition, ovulation induction and in vitro fertilization may increase economic burden on patients, as well as adverse effects on their psychological status.^[16–18] Traditional Chinese medicine (TCM) therapies such as herbal formulas, herbal extracts, and acupuncture have become increasingly important interventions for the prevention and treatment of PCOS according to extensive data analysis.^[19] Numerous meta-analyses have confirmed that TCM therapies are superior to western medicines and surgical treatments with regard to cycle ovulation rate, pregnancy rate, and adverse effects.^[20–26] Numerous clinical studies and experimental data above are conclusive in demonstrating that TCM is proficient in regulating endocrine and metabolic imbalances prevalent in PCOS with the characteristics of multi-components, multi-targets, and multi-pathways, and that the drug safety is high.^[27–32]

However, TCM therapies are diverse and have varying efficacies. Conventional meta-analysis methods can only achieve pairwise direct comparison between single therapies. As a result, the lack of comparative evaluation of multiple TCM combination therapies hinders clinicians in clearly judging their therapeutic value. This limitation is detrimental to the exploration of the maximum advantages of TCM combination therapies and the selection of optimal treatment solutions. Network Meta-analysis (NMA), as an essential tool in evidence-based medicine, is well-suited to address this situation by calculating the comparative effectiveness of multiple therapies for a given disease, ranking them according to direct and indirect evidence, and finding the optimal therapeutic program.^[32–37] Therefore, this study aimed to search for randomized controlled trials of TCM treatments for PCOS infertility in recent years both domestically and internationally. We will utilize NMA to mine optimal treatment solutions for this disease. The objective of this is to provide a reliable evidence-based foundation for clinical application.

2. Methods

2.1. Registration

The protocol for this systematic review and NMA is based on the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocols.^[38] This protocol has been registered with PROSPERO (registration number: CRD42023410217).

2.2. Ethics approval and consent to participate

Ethical approval and patient consent are not required since this is an overview based on published studies.

2.3. Inclusion criteria

2.3.1. Research types

This study aims to collect all randomized controlled trials (RCTs) on the combined Chinese Medicine Protocol for infertility patients with PCOS. There are no language restrictions.

2.3.2. Research participants

The enrolled patients were definitively diagnosed with infertility caused by PCOS, in compliance with the Rotterdam criteria which was proposed by the European Society of Reproduction and Embryology and Autonomous Sensory Meridian Response in 2003.^[39] Such as: ① sporadic ovulation or anovulation; ② clinical evidence of hyperandrogenemia and/or high androgen levels; ③ presence of polycystic ovaries on ultrasound (one or both ovaries having more than 12 follicles with diameters ranging from 2 to 9 mm and/or an ovary volume >10 mL). (Note: the above conditions meet 2, and exclude other diseases such as congenital adrenocortical hyperplasia, Cushing syndrome, etc). The diagnostic criteria for infertility were in line with the definitions outlined by the World Health Organization, Obstetrics and Gynecology,^[40] or Gynecology of Traditional Chinese Medicine,^[41] such as a current and uninterrupted sexual life for over a year without contraception and no conception. There were no limitations concerning the age, duration of illness, or source of the cases of the included patients.

2.3.3. Interventions

Trial group: The combined application of TCM compounds, acupuncture, and other therapies based on TCM theory and meridian and acupoint theory is allowed, with no restrictions on the intervention start time, treatment duration, Chinese herbal selection, or acupoint selection. Control group: conventional western medicines such as Letrozole, Clomiphene, and Metformin, placebo (sham acupuncture), or one of the combined therapies (as used in the trial group).

2.3.4. Outcome indicators

The following outcome indicators were considered: ① clinical pregnancy rate; ② ovulation rate.

2.4. Exclusion criteria

Studies that met the following criteria were excluded: ① studies with unclear or undefined diagnostic criteria. ② Duplicate studies published in multiple languages or formats, such as journals and conferences. ③ Studies that include patients with infertility not solely attributed to PCOS or those with multiple pathologies. ④ Non-RCT trials, animal studies, reviews, systematic evaluations, and meta-analysis. ⑤ Studies that lack complete data and cannot be remedied by contacting the authors. ⑥ Studies that demonstrate noncompliance with intervention. ⑦ Studies that lack endpoint indicators related to infertility with PCOS.

2.5. Retrieval strategy

The search strategy adopted a combination of subject and free words. We will use computers to search 8 databases, including PubMed, Cochrane Library, Web of Science, Embase, China Knowledge Network, Wanfang Database, VIP Database, and Chinese Biomedical Literature Database (SinoMed) from build database until February 28, 2023. Subject terms and free terms were utilized in a joint search, based on the characteristics of each database, and the search formula was developed and adjusted accordingly. The search terms include: “acupuncture,” “Traditional Chinese Medicine,” “polycystic ovarian syndrome,” “infertility,” et al Taking PubMed as an example, the search strategy is provided in Table 1.

Table 1.

PubMed database search strategy.

#1	((((“Medicine, Chinese Traditional”[Mesh]) OR (“Acupuncture Therapy”[Mesh])) OR (((((((((((((Chung I Hsueh[Ti/Ab]) OR (Hsueh, Chung I[Ti/Ab])) OR (Traditional Medicine, Chinese[Ti/Ab])) OR (Zhong Yi Xue[Ti/Ab])) OR (Chinese Traditional Medicine[Ti/Ab])) OR (Chinese Medicine, Traditional[Ti/Ab])) OR (Traditional Tongue Diagnosis[Ti/Ab])) OR (Tongue Diagnoses, Traditional[Ti/Ab])) OR (Tongue Diagnosis, Traditional[Ti/Ab])) OR (Traditional Tongue Diagnoses[Ti/Ab])) OR (Traditional Tongue Assessment[Ti/Ab])) OR (Tongue Assessment, Traditional[Ti/Ab])) OR (Traditional Tongue Assessments[Ti/Ab]))) OR ((((((((((((((((((((((((Acupuncture Treatment[Ti/Ab]) OR (Acupuncture Treatments[Ti/Ab])) OR (Treatment, Acupuncture[Ti/Ab])) OR (Therapy, Acupuncture[Ti/Ab])) OR (Pharmacoacupuncture Treatment[Ti/Ab])) OR (Treatment, Pharmacoacupuncture[Ti/Ab])) OR (Pharmacoacupuncture Therapy[Ti/Ab])) OR (Therapy, Pharmacoacupuncture[Ti/Ab])) OR (Acupotomy[Ti/Ab])) OR (Acupotomies[Ti/Ab])) OR (Chinese medicine[Ti/Ab])) OR (TCM[Ti/Ab])) OR (Acupuncture[Ti/Ab])) OR (warming-needle moxibustion[Ti/Ab])) OR (electro-acupuncture[Ti/Ab])) OR (Electroacupuncture[Ti/Ab])) OR (point injection therapy[Ti/Ab])) OR (water injection[Ti/Ab])) OR (point application therapy[Ti/Ab])) OR (catgut embedment in acupoint[Ti/Ab])) OR (red-hot needling[Ti/Ab])) OR (fire needling[Ti/Ab])) OR (ear point[Ti/Ab])) OR (auricular acupoint[Ti/Ab]))) OR (“Moxibustion”[Mesh])
#2	(“Polycystic Ovary Syndrome”[Mesh]) OR ((((((((((((((Ovary Syndrome, Polycystic[Ti/Ab]) OR (Syndrome, Polycystic Ovary[Ti/Ab])) OR (Stein-Leventhal Syndrome[Ti/Ab])) OR (Stein Leventhal Syndrome[Ti/Ab])) OR (Syndrome, Stein-Leventhal[Ti/Ab])) OR (Sclerocystic Ovarian Degeneration[Ti/Ab])) OR (Ovarian Degeneration, Sclerocystic[Ti/Ab])) OR (Sclerocystic Ovary Syndrome[Ti/Ab])) OR (Polycystic Ovarian Syndrome[Ti/Ab])) OR (Ovarian Syndrome, Polycystic[Ti/Ab])) OR (Polycystic Ovary Syndrome 1[Ti/Ab])) OR (Sclerocystic Ovaries[Ti/Ab])) OR (Ovary, Sclerocystic[Ti/Ab])) OR (Sclerocystic Ovary[Ti/Ab]))
#3	(“Infertility, Female”[Mesh]) OR (((((((((((Female infertility[Ti/Ab]) OR (Female Infertility[Ti/Ab])) OR (Sterility, Postpartum[Ti/Ab])) OR (Postpartum Sterility[Ti/Ab])) OR (Subfertility, Female[Ti/Ab])) OR (Female Subfertility[Ti/Ab])) OR (Sub-Fertility, Female[Ti/Ab])) OR (Female Sub-Fertility[Ti/Ab])) OR (Sub Fertility, Female[Ti/Ab])) OR (Sterility, Female[Ti/Ab])) OR (Female Sterility[Ti/Ab]))
#4	“randomized controlled trial”[Mesh] OR “RCT”[Mesh] OR “randomly”[Ti/Ab] OR “randomized”[Ti/Ab] OR “controlled clinical trial”[Ti/Ab] OR “clinical trial”[Ti/Ab]
#5	#1 AND #2 AND #3 AND #4

2.6. Literatures screening and data collection

Two researchers independently conducted a thorough literature review, extracted pertinent information, cross-checked for accuracy, and established a comprehensive database. Disagreements regarding the literature were resolved through consultation with other qualified researchers. EndNote (V.X.9.2) software was used for screening literatures: ① delete duplicate literature; ② read the title and abstract to exclude literature; ③ read the full text for secondary screening according to inclusion and exclusion criteria. (Full text information could be obtained by e-mail and other means of authors if needed during the screening process.)

Data collection: ① title, author, publication time, etc; ② baseline characteristics and interventions; ③ elementary risk of bias evaluation indicators; ④ outcome indicators measurements data.

2.7. Assessment of the risk of bias

Risk of bias was evaluated according to the quality evaluation tool (RoB 2.0) recommended in the Cochrane Handbook 5.1.0.^[42] The assessment was conducted by 2 investigators independently, cross-checked, and any discrepancies were resolved through discussion with the other investigators. Evaluation contents: random sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other possible biases. According to the criteria, each item was judged as “low risk,” “high risk,” or “unclear risk.”

2.8. Statistical analysis

The outcome indicators in this study were all dichotomous variables, and utilized odds risk (OR) as the effect size, and its 95% confidence interval (CI) was calculated. Taking into account inter-study heterogeneity, the data were analyzed using a random effects model.^[43] NMA was conducted using Stata 16.0 software by using the network package command. The evidence network diagram depicts the number of patients who received the intervention, with larger dots indicating more significant numbers, while the thickness of the line between 2 interventions represents the number of studies included.^[44] The area under curve of surface under the cumulative ranking (SUCRA) was expressed as a percentage (range, 0–100%) to explain the superiority or inferiority of each intervention, and the percentage size is proportional to the efficacy of the intervention.^[45–47] In addition, a “comparison-corrected” funnel plot was used to assess the publication bias and small sample of included studies.^[48]

3. Results

3.1. Results of study identification and selection

Through a rigorous search of diverse databases, a total of 4168 articles were initially identified. This consisted of 978 articles sourced from China Knowledge Network, 875 articles retrieved from Wanfang Database, 661 articles obtained from VIP Database, 1020 articles obtained from SinoMed, 57 articles retrieved from Cochrane Library, 293 articles obtained from Embase, 52 articles obtained from PubMed, and 232 articles obtained from Web of Science. Following meticulous evaluation of the title and abstract, 1699 articles were excluded, leaving only 28 RCTs^[49–76] for further assessment and meta-analysis. The screening process and results for this stage are depicted in Figure 1.

Figure 1.

Literature search process.

3.2. Characteristics of included studies

We finally included 28 RCTs were published in Chinese, except for 3^[61–63] three-arm experiments, all double-arm experiments with a total of 2771 participants, 1300 of whom were assigned to the intervention group and 1471 to the control group. The publication time of the studies was distributed from 2012 to 2022. There were 11 different interventions were involved, including moxibustion + herbal,^[49–54] electro-acupuncture + herbal,^[55] ear point pressure + herbal enema + herbal,^[56] fire needling + herbal,^[57] catgut embedment in acupoint + herbal,^[58] acupoint application + herbal,^[59] acupuncture + herbal,^{[60–64,70–76]} acupuncture and moxibustion + herbal,^[65–69] western medicine,^{[49–55,58,65,70–76]} herbal,^{[56,57,59,61–64,66–69]} and acupuncture.^[60–63] The outcome of the included studies was the pregnancy rate or the ovulation rate. The basic information of the included studies is shown in Table 2.

Table 2.

List of basic features included in the study.

Study	Test group				Control group				Treatment (menstrual cycle)	Outcome measure
	Intervention	Cases	Age (years)	Course of disease(years)	Intervention	Cases	Age (years)	Course of disease(years)
Jia 2012[49]	MH	60	28.35 ± 4.16	3.89 ± 1.76	W	60	24.49 ± 4.02	3.79 ± 1.93	3	①②
Xu 2014[50]	MH	128	28.27 ± 4.56	3.72 ± 1.59	W	122	28.41 ± 5.02	3.78 ± 1.63	3	②
Zhu 2019[51]	MH	30	26.24 ± 5.20	3.48 ± 1.63	W	30	26.58 ± 4.82	3.35 ± 1.72	3	①②
Si 2016[52]	MH	43	27.90 ± 5.00	4.00 ± 2.10	W	43	28.40 ± 4.80	3.70 ± 2.00	3	②
Lu 2019[53]	MH	50	28.40 ± 3.70	4.40 ± 0.70	W	50	28.80 ± 3.90	4.20 ± 0.80	3	①②
Liu 2107[54]	MH	43	59.70 ± 12.80	–	W	43	59.70 ± 12.80	–	6	②
Peng 2018[55]	EAH	60	26.41 ± 2.54	–	W	60	25.14 ± 3.85	–	3	②
Zhu 2016[56]	EHH	30	27.35 ± 3.26	4.10 ± 1.50	H	28	28.38 ± 4.19	4.20 ± 2.10	3	①②
Geng 2020[57]	FH	67	31.00 ± 7.00	3.24 ± 0.74	H	67	31.00 ± 7.00	3.22 ± 0.76	40 days	①②
Zhu 2020[58]	CH	30	26.37 ± 5.31	3.63 ± 1.58	W	30	26.47 ± 5.62	3.41 ± 1.49	3	①②
Shi 2022[59]	AAH	44	29.56 ± 1.09	–	H	44	29.36 ± 1.12	–	3months	②
Zhou 2022[60]	AH	40	26.87 ± 2.19	3.15 ± 0.48	A	40	27.13 ± 2.26	2.82 ± 0.37	3	①②
Ren 2022[61]	AH	80	28.00 ± 5.00	4.03 ± 0.59	A	80	29.00 ± 5.00	4.05 ± 0.61	3 months	①②
					H	80	17.00 ± 5.00	3.95 ± 0.62	3 months	①②
Lei 2021[62]	AH	70	31.00 ± 4.00	3.50 ± 0.90	A	70	32.00 ± 5.00	3.50 ± 0.6	3	①②
					H	70	31.00 ± 4.00	3.40 ± 0.70	3	①②
Yang 2022[63]	AH	30	30.20 ± 3.21	3.40 ± 0.35	A	30	29.90 ± 3.05	3.40 ± 0.37	3	①②
					H	30	29.8.0 ± 2.87	3.30 ± 0.29	3	①②
He 2021[64]	AH	33	25.57 ± 0.75	1.64 ± 0.03	H	33	25.72 ± 0.73	1.64 ± 0.03	3	②
Jiang 2015[65]	AMH	40	29.00 ± 3.00	3.80 ± 2.10	W	40	28.00 ± 3.00	4.10 ± 1.90	3	②
Zhong 2019[66]	AMH	63	33.00 ± 5.00	1.50 ± 0.62	H	63	33.00 ± 5.00	1.53 ± 0.60	6	①②
Peng 2022[67]	AMH	32	29.91 ± 4.54	–	H	32	28.47 ± 3.10	–	–	①②
Cui 2015[68]	AMH	33	31.38 ± 4.87	1.54 ± 1.01	H	33	33.45 ± 3.75	1.43 ± 0.79	6	②
Qiao 2012[69]	AMH	30	–	–	H	30	–	–	6	①②
Wang 2013[70]	AH	28	30.30 ± 2.70	7.00 ± 2.70	W	28	31.00 ± 3.10	7.00 ± 2.90	3	②
Shuai 2017[71]	AH	33	28.59 ± 1.57	2.55 ± 1.82	W	32	29.21 ± 1.97	2.32 ± 1.51	3 months	①②
Luo 2018[72]	AH	30	29.63 ± 2.69	3.87 ± 1.73	W	30	30.21 ± 3.19	3.48 ± 1.77	3	②
Li 2021[73]	AH	33	29.63 ± 4.02	5.38 ± 1.24	W	33	29.53 ± 4.09	5.39 ± 1.22	3	①②
Yu 2020[74]	AH	55	26.56 ± 6.11	2.38 ± 0.89	W	55	25.39 ± 4.89	3.59 ± 0.75	2–6	②
Gao 2017[75]	AH	45	28.89 ± 2.50	2.81 ± 1.30	W	45	29.05 ± 2.32	2.98 ± 1.26	–	①②
Li 2019[76]	AH	40	33.10 ± 1.60	4.30 ± 0.50	W	40	32.80 ± 1.70	4.10 ± 0.70	3 months	①②

Notes: –: not reported, MH: moxibustion + herbal, EAH: electro-acupuncture + herbal, EHH: ear point pressure + herbal enema + herbal, FH: fire needling + herbal, CH: catgut embedment in acupoint + herbal, AAH: acupoint application + herbal, AH: acupuncture + herbal, AMH: acupuncture and moxibustion + herbal, W: western medicine, H: herbal, A: acupuncture, ①: pregnancy rate, ②: ovulation rate.

3.3. Quality assessment of the included studies

We assessed the risk of bias for each study using the Cochrane RoB–2.0 tool. The generation method of specific random sequences was reported by 20 studies, of which 17 studies^{[49,50,53,54,57,59,61–66,68,70,72–74]} used random number table for random allocation, one study^[76] used lottery for random allocation, rated as “low risk.” One study^[69] used sequence of medical visits, one study^[52] used treatment methods, both in non-randomized grouping, rated as “high risk.” Meanwhile, 8 studies^{[51,55,56,58,60,65,67,71]} mentioned randomized grouping, but failed to specify the specific way of generating the randomized sequence, rated as “unclear risk.” None of all studies^[49–76] mentioned whether allocation concealment was performed was evaluated as “unclear risk.” None of all studies^[49–76] mentioned the use of blinding for patients, intervention implementers, data analysts, and outcome assessors, which was evaluated as “high risk.” In 4 studies^{[61,62,68,72]} detailed information concerning shedding rates and reasons for missing cases were reported, the missing data were insufficient to affect the predicted intervention effect therefore classified as “low risk”; and the remaining were rated as “unclear risk.” None of all studies^[49–76] mentioned clinical study registration information, and the complete study protocols were not available hence classified as “unclear risk.” There was no mention of other biases within any of the studies,^[49–76] which were categorized as “unclear risk.” The evaluation table for risk of bias is depicted in Figure 2.

Figure 2.

Literature quality evaluation results.

3.4. Evidence network

The reticulation diagram depicts individual interventions through the use of dots, the size of each dot corresponds to the sample size of the included cases, and the thickness of the lines signifies the sample size of the studies analyzed for inclusion in the analysis. Additionally, 2 connected dots indicate a direct comparison between the interventions, while unconnected dots represent indirectly analyzed comparisons through NMA. All the studies^[49–76] presented data on pregnancy rates, which included 11 different interventions and the network relationship was centered on western medicine therapy (Fig. 3A). In addition, 18 studies^{[51,53,56–58,60–63,65–67,69,71,75,76]} reported the ovulation rate, involving 9 interventions and the network relationship was centered on herbal therapy (Fig. 3B).

Figure 3.

Evidence network relationship figure.

3.5. Network meta-analysis

3.5.1. Pregnancy rate

The NMA of the yielded studies resulted in 55 pairwise comparisons. By combining the OR and 95% CI, the NMA results revealed that MH, FH, AH had better intervention effects compared to AMH, A, H, and W. Additionally, EAH showed better efficacy than W, and AAH showed better efficacy than H. All of the above differences were statistically significant (P < .05). No statistically significant differences in the comparisons (P > .05) were observed between the other interventions, as demonstrated in Figure 4.

Figure 4.

Network meta-analysis of the pregnancy rate.

3.5.2. Ovulation rate

The NMA of the included studies yielded 36 pairwise comparisons. Combining the OR and 95% CI, the results of NMA showed that MH had better efficacy than W and EHH; compared with W, H, the intervention effect of EHH, and AMH was better. Furthermore, FH and AH showed better efficacy than W, H, and A. All of the above differences were statistically significant (P < .05). Conversely, there were no statistically significant differences in the comparisons (P > .05) between the other interventions, as illustrated in Figure 5.

Figure 5.

Network meta-analysis of the ovulation rate.

3.6. SUCRA probability ranking

3.6.1. Pregnancy rate

According to SUCRA results, MH appears to be the most efficacious intervention. The SUCRA probability ranking, from highest to lowest, is as follows: MH (84.1%) > FH (77.1%) > AH (67.0%) > CH (66.3%) > AAH (64.5%) > EAH (58.9%) > EHH (58.4%) > AMH (28.1%) > A (22.9%) > W (13.1%) > H (9.5%), as depicted in Figure 6.

Figure 6.

SUCEA of the pregnancy rate.

3.6.2. Ovulation rate

According to SUCRA results, FH appears to be the most effective intervention, with the SUCRA probability ranking highest to lowest as follows: FH (86.5%) > AH (78.7%) > EEH (72.0%) > AMH (65.0%) > MH (53.3%) > CH (48.9%) > W (20.2%) > H (15.1%) > A (10.3%), as shown in Figure 7.

Figure 7.

SUCRA of the ovulation rate. SUCRA = surface under the cumulative ranking.

3.7. Publication bias

In the comparison-corrected funnel plot, the dots in the funnel chart of different colors represent direct comparisons between different therapies, with the number of dots indicating the number of studies. Regarding the pregnancy rate outcome indicators, which consist of 12 distinct comparison groups for the interventions, most of the dots in the funnel plot were symmetrically distributed on the vertical line and its corresponding sides. However, the scatterplot of the points indicates the possibility of some degree of publication bias (Fig. 8A). Regarding the ovulation rate outcome indicators, which comprise 8 distinct comparison groups for the interventions, most of the dots in the funnel plot were symmetrically distributed on the vertical line and its corresponding sides. However, one group stands out as being further away from the regression line, suggesting the presence of some degree of publication bias (Fig. 8B).

Figure 8.

The funnel plot of included studies.

4. Discussion

4.1. Results analysis

The findings of this study demonstrate that moxibustion + herbal, fire acupuncture + herbal, acupuncture + herbal, electroacupuncture + herbal, and acupoint application + herbal have some degree of advantages when exhibited improved the clinical pregnancy rate among PCOS-related infertility in comparison to monotherapy involving acupuncture, herbal medicine, and western medicine. Moreover, ear point pressure + herbal Enema + herbal, acupuncture and moxibustion + herbal, fire acupuncture + herbal, and acupuncture + Chinese herbal medicine also have some degree of advantages when enhanced ovulation rate in PCOS-related infertility compared to monotherapy comprising of acupuncture, herbal, and western medicine. When combined TCM therapy was used as alternative therapies, the cumulative probability ranking suggested that the top 3 efficacies in terms of improving pregnancy rate were moxibustion + herbal, fire acupuncture + herbal, and acupuncture + herbal, and the top 3 efficacies in terms of improving ovulation rate were fire acupuncture + herbal, acupuncture + herbal, ear point pressure + herbal enema + herbal. In summary, it is evident that the optimal treatment programs varied based on the outcome indicators, and there was no optimal treatment program for each outcome indicator.

In the TCM, there is no specific term for PCOS, but based on its clinical symptoms, it can be classified as “amenorrhea,” “lower abdominal masses,” “infertility,” “scanty menstruation,” and so on. Due to the complexity of PCOS, Chinese medicine experts have varying opinions on its etiology and pathomechanism, but generally agree that it can be treated through regulating the functions of the kidney, spleen, and liver. Previous studies, when statistically analyzing the TCM syndromes of PCOS, showed that kidney deficiency is the main syndrome of PCOS and has the highest frequency of occurrence.^[53] On one hand, kidney essence insufficiency can lead to the inability to nourish the tiangui and affect the development and maturation of the ovum. On the other hand, kidney essence insufficiency can involve the functions of the liver and spleen, generating phlegm-dampness and blood stasis, hinder the function of the Chongren, uterus or skin, causing difficulty with the discharge of ovums and symptoms such as hirsutism, obesity, and acne. It can be seen that this disease is based on kidney deficiency, involving the liver and spleen, phlegm-dampness and blood stasis as the target, interacting with each other, disrupting the dynamic balance of the reproductive axis of “kidney – tiangui – Chongren—uterus” and causing metabolic dysfunction of the body, thus affecting women’s pregnancy and impregnation in addition morbidity.^[77]

As an adjunctive therapy, TCM can offer personalized treatment from a holistic perspective combining pattern differentiation and treatment. Recent research has indicated that TCM effectively manages PCOS through herbal formulations and acupuncture.^[78] Modern pharmacological examination has validated the efficacy of many herbal monomers. A series of herbals that tonify the kidney, fortify the spleen, dissolve phlegm, dry dampness, invigorate blood and dissolve stasis, can regulate hormone levels, metabolic indexes, and localized ovarian perfusion, which enhance the ovarian and uterine microenvironment, thicken endometrium, promote follicular growth and follicle expulsion, and ultimately improve pregnancy outcomes and relieve other symptoms.^[79–82] The most common cycle-regulating therapy in TCM can regulate menstruation and promote ovulation through the patterns of qi and blood and yin-yang changes and the administration of medication at different stages.^[83] Its mechanism is being gradually explored. Xu et al^[84] discovered that adjusting the weekly cycle method can improve local microcirculation in PCOS patients, increasing blood supply to the uterus and ovaries, promoting follicle development, and ovulation. Shen et al revealed that this method can also reduce serum levels of LH, LH/FSH, hirsuteness, and acne incidence while improving ovulation rate.^[85] Acupuncture treatment can support the body, improve kidney function, strengthen the spleen, and regulate qi and blood flow by selecting appropriate acupoint pairings, such as Sanyinjiao (SP6), Guanyuan (RN4), Zigong (EX-CA1), Zhongji (RN3), and Qihai (RN6).^[86] Additionally, acupuncture can be tailored to address specific symptoms by adding or subtracting acupressure points. Biologically, acupuncture can regulate hypothalamic function, pituitary neurotransmitter release, and ovarian function, which in turn adjusts hormone levels and the uterine microenvironment.^[87,88] Animal experiments have illustrated that electroacupuncture can stimulate homeostasis of the hypothalamic–pituitary–ovarian axis in normal female rats and regulate levels of gonadotropin-releasing hormone, luteinizing hormone, follicle-stimulating hormone, estradiol, and progesterone.^[89] Furthermore, acupuncture may improve follicular maturation in patients with PCOS by modulating ovarian innervation and decreasing elevated levels of nerve growth factor in the ovaries, in order to achieve the desired therapeutic effect.^[88,90]

In summary, TCM has shown significant efficacy in treating PCOS. Although there are numerous therapies with significant efficacy, the adverse effects of the drugs limit their clinical applicability. Specific molecular studies and toxicology tests of Chinese herbs are still lacking and require further clarification to provide effective targeted therapy for PCOS. Nonetheless, the interventions mentioned above provide some reference for clinical application and research. Please note that the probability ranking results are for reference purposes only.

4.2. Strengths and limitations of this study

Strengths: The literature screening was focused on minimizing differences by focusing on diagnostic, inclusion, exclusion, and efficacy criteria of the included studies. This study selected commonly used TCM therapies to treat PCOS infertility, which expands the range of treatment options, compares interventions, and explores targeted and efficient therapies.

Limitations: The number of outcome indicators in this paper is limited, and the number of RCTs included under different interventions is small. The overall quality of the RCTs included in this paper is not high, and most of the literature did not use allocation concealment methods and record case shedding, which may impact the evaluation of results. The types of Chinese medicine symptoms, the start time of the interventions, and the intervention course of the included studies are inconsistent, which may result in differences in outcome measurements. The included studies examined different types of Chinese medicine symptoms about PCOS infertility, and subgroup analyses could not be performed due to limitations in the literature.

4.3. Implications for future research

4.3.1. Trial design

A majority of the studies included had small sample sizes, and there was limited utilization of blinding methods with concealed allocation. Blinding is a critical component in reducing potential biases in results, especially for subjectively measured outcomes. While achieving complete double-blinding or triple-blinding designs in acupuncture studies may be challenging, blinding of acupuncture subjects or data collectors or evaluators can help minimize subjective factors and result biases. Additionally, partially randomized patient preference clinical trials may be an effective approach to address these limitations.^[91,92]

4.3.2. Adverse reactions

Adverse events play a critical role in clinical trials as they provide important safety information for evaluation. To ensure objective and accurate data collection, future studies should document adverse events in accordance with the clinical protocol.

4.3.3. In terms of diagnostic and efficacy criteria

The included studies lacked uniform diagnostic reference criteria. In future research, it would be beneficial to establish validated diagnostic criteria based on existing literature or guideline recommendations to enhance measurement accuracy and promote comparability with other studies in the field.

In conclusion, future research should adhere to the requirements of controlled trials, rigorously control diagnostic criteria, reference previous studies or guideline outcome indicators for assessment, design and implement large-scale trials, and faithfully record adverse reactions and safety assessments to achieve higher-quality RCTs.

5. Conclusions

This research has demonstrated that the combined TCM therapy was effective in improving clinical pregnancy and ovulation rate when used as an alternative treatment for PCOS infertility. Moxibustion combined with herbal may become the optimal therapy to improve pregnancy rate, and fire acupuncture combined with herbal may become the optimal therapy to improve ovulation rate, which has some clinical reference value. In clinical application, the results of this study can be appropriately referred to and appropriate therapies can be selected according to the specific conditions of patients. However, it is important to acknowledge that further research is necessary to verify these findings via large samples RCTs that are of high quality and standardized in nature.

Author contributions

Conceptualization: Yun-Peng Deng, Yan-Li Zhou, Xu-Guang Yang.

Data curation: Yun-Peng Deng, Xu-Guang Yang.

Investigation: Yan-Li Zhou, Teng-Teng Wei, Guang-Shuai He, Zhi-Xin Zhu, Shu-Ning Zhang, Mei-Jun Liu, Jing-Jing Xue, Wei-Xing Zhang.

Methodology: Yun-Peng Deng, Xu-Guang Yang.

Software: Yun-Peng Deng, Teng-Teng Wei, Guang-Shuai He, Zhi-Xin Zhu, Shu-Ning Zhang, Mei-Jun Liu, Jing-Jing Xue, Wei-Xing Zhang.

Validation: Teng-Teng Wei, Guang-Shuai He, Zhi-Xin Zhu, Shu-Ning Zhang, Mei-Jun Liu, Jing-Jing Xue, Wei-Xing Zhang.

Visualization: Yan-Li Zhou.

Writing – original draft: Yun-Peng Deng.

Writing – review & editing: Yun-Peng Deng, Xu-Guang Yang.