Acupuncture-Related Therapies for Ovulatory Disorders: A Systematic Review and Bayesian Network Meta-Analysis Protocol

Background(cid:0)Ovulation disorders (ODs) are major cause of infertility. Many clinical studies have shown that acupuncture is potentially useful in the treatment of ODs. In this context, the present study aims to evaluate the effectiveness and safety of different acupuncture therapies through systematic review and Bayesian network meta-analysis (NMA). Methods(cid:0) A comprehensive literature search 2021 will be carried out in the following National China biomedical database Randomized controlled trials (RCTs) meeting the eligibility criteria based on PICO elements will be included. The main outcomes will be ovulation rate, pregnancy rate and changes in female reproductive hormones. The secondary outcome will be the prevalence of adverse events directly related to acupuncture. Cochrane bias risk assessment tool (RoB 2.0) will be used to evaluate the quality of the selected literature. Stata, Aggregate Data Drug Information System (ADDIS)and BUGS will be used to manage the data.


Background
Normal ovulation requires the proper function of the hypothalamus-pituitary-gonadal axis [1]. Any organic lesion or function of this axis can cause temporary or long-term ovarian dysfunction, leading to abnormal ovulation. Ovulation disorders (ODs) have multiple causes including abnormalities of the central nervous system [2], hypothalamus [3], pituitary and ovaries, polycystic ovary syndrome [4,5,6], luteinized unruptured follicle syndrome [7]. Additionally, disorders affecting other endocrine systems, such as thyroid and adrenal cortex dysfunctionas well as selected systemic diseases (e.g, severe malnutrition), can affect ovarian function and lead to ovulation dysfunction. Acupuncture is a traditional medical treatment in China. Several studies have shown that acupuncture treatment of ODs has a signi cant, long-lasting bene cial effects [8,9] and that the adverse reactions are limited [10,11,12,13].
However, other studies questioned that the effectiveness of acupuncture in this setting remains to be demonstrated [14] and that the quality of evidence-based clinical observations is low.
Currently it is not possible to recommend the clinical use of acupuncture in the treatment of ovulatory disorders. Nor it is possible to recommend the best treatment scheme. This is due to several reasons: the variety of acupuncture-related therapies, the different emphasis on e cacy, the inconsistent conclusions of the studies and the lack of direct comparisons between different acupuncture-related therapies.
Therefore, in addition to the conventional meta-analysis method, the present study uses the network meta-analysis method to compare the e cacy of a variety of acupuncture related methods on ovulation rate, pregnancy rate and changes in reproductive hormones of patients with ovulation disorders. The aim of this study is to provide evidence-based medicine for a clinical selection of the best acupuncture related treatment scheme to improve several reproductive indicators in patients with ODs ( Figure 1).
Network meta-analysis (NMA) [15,16] is a technique suitable for comparing multiple treatments simultaneously in a single analysis. It combines direct and indirect evidence within a network of RCTs by making cross-comparisons. It can also compare the e cacy of at least two interventions by combining data from different trials and by ranking all the included interventions. Therefore, it is also called mixed treatment comparison meta-analysis or multiple treatments meta-analysis. The purpose of this paper is to compare the effects of various acupuncture-related measures on ovulation rate and pregnancy rate of patients with ODs by using the network meta-analysis method, to provide a reference for clinical application ( Figure 2)

Eligible criteria Normative references
The results of this study will be reported following the National Institute for health and care excellence (NICE) network meta-analysis reporting standard.

Research type
Published RCTs of acupuncture-related therapy for ovulation disorders

Research objects
Patients with ODs refer to the disease types published by up-to-date(https://www.uptodate.com/) .Combined with World Health Organization (WHO) standards, ODs are divided into four categories (a) WHO I type ovulation disorder: there is hypothalamus or pituitary failure with low level of endogenous estrogen, low level of FSH and LH;(b) WHO II type: normo-gonadotropic, normo-estrogenic anovulation, with possible imbalance in FSH and LH levels, common in patients with polycystic ovary syndrome and hyperprolactinemia;(c) WHO III patients: ovarian failure, FSH and LH levels are increased, estrogen levels are decreased; (d) ODs different from those included in the above categories. Examples are luteinized unruptured follicle syndrome (LUFS). Patients with severe medical conditions were excluded. See Table  1 for detailed retrieval strategy.

Exclusion criteria
The following are considered not to meet the inclusion criteria: (a) Abstracts, conference papers, dissertations, case reports, animal mechanism studies; (b) Repetitive publications; (c) Publications in which there are no de nite diagnostic criteria nor clear criteria to evaluate the e cacy of treatments; (d) Absence of a widely accepted method of randomization; (e) The intervention does not include acupuncture related treatment; (f)Articles that are not available after having requested them to the authors; (i) Articles published belonging to the same study. Refer to Table 2 for the draft quali cation criteria for inclusion and exclusion.

Types of interventions
The treatment group is formed by women treated with all kinds of acupuncture-related therapies, including electroacupuncture, auricular acupuncture, warm acupuncture, auricular point sticking and pressing, moxibustion, acupoint catgut embedding. The control group is formed by women treated with Western medicine or placebo, The frequency of acupoint use and the frequency of intervention measures are counted and displayed by histogram. We show the possible patterns of different intervention methods on the acupuncture points related to ovulation disorders, Using Cytoscape (version 3.80) to construct the network framework diagram of "acupuncture-method-acupoint-ODs type" (Figure3), and reveal the degree of acupuncture points for different diseases and different acupuncture programs.

Outcome measures
Main outcome measures: Ovulation rate, pregnancy rate and changes in reproductive hormones. Different ODs will be discussed separately.
Secondary indicators The prevalence of adverse events directly related to interventions will be considered as secondary outcome.

Literature retrieval strategy
PubMed, EMBASE, Cochrane Library, CBM, CNKI, Wan Fang, and VIP database will be searched by computer. All the available literature will be searched until June 2021. In addition, the references of the included literature are traced to supplement the relevant literature. The retrieval is carried out by combining subject words with free words.

Literature screening and data extraction
Two researchers will independently screen, extract, and cross-check the literature. In case of disagreement, both parties shall reach an agreement through discussion or a third researcher shall assist in the exclusion process. Data will be extracted by a uni ed data extraction table, including title, author, sample size, average age, gender, course of the disease, intervention measures, course of treatment, outcome indicators. The selection process will be performed in the PRISMA owchart (see checklist in Additional le 1), as shown in Figure 4.

Bias risk assessment of included studies
Two researchers will evaluate the included studies according to the bias risk assessment tool recommended in Cochrane Handbook 5.1. (a)Random sequence generation; (b)Allocation concealment; (c)Blinding of participants and personnel; (d)Blinding of outcome assessment; (e)Incomplete outcome data; (f)Selective outcome reporting; (g) Assessing risk of bias from other sources. If the two sides do not agree, they will discuss with the third researcher to reach an agreement.

Data analyses
Quality assessment The included literature will be assessed according to the recommended levels of evaluation development and evaluation (grading) guidelines to assess the quality of evidence. An assessment and grading of the quality of the included literature will be performed according to the recommendations of the guideline (quote). The guidelines classify the quality of evidence as high, medium, and low. The entire study ow is reported in Figure 5. Funnel plots and the Egger regression test will be performed to examine potential publish bias. In addition, sensitivity analysis will be carried out by sequentially deleting trials to check the stability of the primary outcomes.

Subgroup analysis
If the relevant studies are su cient, the following characteristics will be considered for subgroup analysis: (a) RCT studies on the same type of ovulation disorder but with different diseases or causes; (b) RCT studies on different acupuncture intervention methods for the same disease; (c) RCT studies on different acupuncture points for the same treatment method.

Pairwise meta-analysis
The premise of Pairwise Meta-analysis is the law of similarity, which means that the meta-analysis can be carried out only when the included studies meet a certain degree of similarity, we state that no less than 3 studies of the same interventions and outcome indicators can make sense in pairwise metaanalysis. By using Stata 14.0, odds ratio (OR) and 95% con dence interval (CI) will be adopted. I 2 test will be applied to assess heterogeneity and select a model. A random-effect model will be adopted if I 2 >50% while a xed-effect model will be adopted for I 2 values <50%. Before selecting a model, sensitivity analysis will be performed properly to delete high-heterogeneity studies. Employing Begg's testing, public bias will be estimated by symmetry of the funnel plot if more than 10 pairwise comparison studies are included.
Bayesian analysis WinBUGS 1.4.3 software is used for statistical analysis. When drawing the evidence network diagram, the test with three arm trials or more is divided into all the double arm trials of all possible combinations [17]. The average effect of all pairing comparisons and the contribution to the whole network are calculated.
Next, we make a comparison-correction funnel chart to evaluate whether the included studies have a small sample effect [15]. Inconsistency factors (if) and 95% CI are used to evaluate the consistency of each closed loop. The Markov chain Monte Carlo (MCMC) random effect model will be analyzed through the WinBUGS program, using four chains to simulate and set the number of iterations to 50000 [18]. The area under the cumulative ranking probability graph (Sucra) will be drawn to predict the e cacy ranking of each treatment measure [19].

Ethics and informed consents
Because this is a systematic literature research program, ethical approval can be skipped. At the same time, the agreement has been registered in INPLASY https://inplasy.com/ . The registration ID is INPLASY202160078.

Patient and Public Involvement
This subject of this study is the second article, and no patients or the public participated in this study.

Strengths and limitations of this study
This protocol permitted the design of a multi-dimensional network map of research involving acupuncture and moxibustion for the treatment of different ovulation disorders (ODs).
By applying network analysis and Constructing the network of "acupuncture-methods-acupoints-ODstype", this protocol comprehensively evaluates the e cacy, safety, and best treatment combinations involving needle massage for the treatment of ODs.
This protocol applied the WHO classi cation system to classify ODs to provide a research report with contemporary clinical signi cance. This work provides signi cant guidance for researchers engaged in acupuncture and evidence-based moxibustion research.
Due to the global and multilingual aspects of this study, consensus discussion will be limited with only a select number of individuals.

Discussion
ODs are an important cause of infertility [20], with a considerable impact on the psychological health of the women affected as well as on their familial relationships. The wide application of acupuncture and moxibustion in clinical practice leaded to a rapid expansion of the research on their effects in the treatment of ODs [21,22,23,24]. The therapeutic results of complementary and alternative medicine are promising. Even though the mechanism of action of acupuncture is still uncompleted understood, modern technology has shown that acupuncture can regulate the neuroendocrine-immune system [25], with potential bene t for patients with ODs. However, the clinical application of acupuncture and moxibustion is complex. The increasing knowledge on acupuncture and moxibustion and the development of new instruments make the therapeutic decision di cult, since it has not yet clearly determined which is the best treatment.
It is well known that there are many different diseases causing ovulation disorders and various treatment approaches. Therefore, we divided ODs into four categories according to the World Health Organization (WHO) standards with little modi cations. To our knowledge, so far there is only one published systematic review and Bayesian analysis on acupuncture in the treatment of ODs, and it is limited to PCOS [26]. The standardized evaluation of acupuncture treatment of ovulation disorders, including acupoints, treatment time, intervention frequency, and the understanding of different types of diseases still need to be further studied. Therefore, how to develop a reasonable and effective treatment plan is an urgent, still unresolved problem. This is the reason for which we decided to carry out a network metaanalysis. If enough data will be available, we will also perform subgroup analysis. The results of the program will be published in relevant journals and updated quickly when and if needed.   y If you answer NO to any of these questions, the citation/study will be excluded. All other full-text articles will be included.  PRISMA ow diagram of the study selection process Data analysis ow chart