1 Introduction
Primary dysmenorrhea (PD) is a common and often debilitating gynecological disease that usually begins in adolescence after the establishment of ovulatory cycles.[1,2] [3–5] [6,7] [8–10] [11,12]
Now pharmacological interventions including nonsteroidal antiinflammatory drugs (NSAIDs) and hormonal contraceptives have been the mainstay of treatment for PD. NSAIDs have the main effect of immediate analgesia, however, they lack of long-term efficacy. Moreover, long-term use of NSAIDs may increase the incidence of adverse reactions, especially in digestive system and central nervous system, such as indigestion, headache, drowsiness, and poor adherence to treatment.[13–15] [16,17] [18]
Complementary and alternative therapies include exercise, acupuncture, moxibustion, Chinese herbal medicines, behavioral interventions, topical heat, dietary supplements, and so on.[19] [20] [21] [22] [22]
There are so many complementary and alternative therapies for PD and their efficacy has been assessed by several systematic reviews. But the traditional pairwise meta-analyses only evaluated the direct comparison of pair-wised drugs and conflicting interpretation of results also existed among different studies. Therefore, the objective of this network meta-analysis is to compare the complementary and alternative therapies in terms of the efficacy and safety for the treatment of PD to better guide clinical practice, the management in women and health policies.
2 Objective
This network meta-analysis (NMA) aims to evaluate the current evidence for the efficacy and safety of complementary and alternative therapies for the PD.
3 Methods and analysis
Our protocol of this NMA was registered on the PROSPERO international prospective register of systematic reviews (ID=CRD42018107763). We have prepared our NMA protocol according to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines.[23] [24]
3.1 Eligibility criteria
This study is a systematic review with NMA of RCTs on complementary and alternative therapies in PD. All relevant RCTs using complementary and alternative therapies for PD will be included. Quasi-RCTs will be excluded such as those allocating by medical record number. The specific population, interventions, comparators, outcomes criteria are as follows.
3.1.1 Population
The population included women after adolescence with PD diagnosed according to various criteria (e.g., Primary Dysmenorrhea Consensus Guidelines)[1]
3.1.2 Interventions and comparators
RCTs of complementary and alternative therapies used in patients with PD will be included, regardless of whether the antimicrobials were tested between against placebo/control intervention or themselves (head-to-head).The complementary and alternative therapies include exercise, acupuncture, moxibustion, Chinese herbal medicines, behavioral interventions, topical heat, dietary supplements regardless of disposal method, and duration.
3.1.3 Outcomes
The primary outcomes of our interest are pain intensity (the most painful day or average pain intensity on days the pain was experienced) and pain duration measured in hours. The pain intensity is measured by validated tool, such as the VAS. The VAS is a valid and reliable tool to determine the severity of pain, from 0 (no pain) to 10 (intolerable pain). The secondary outcomes are quality of life, clinical effective rate, and adverse events. Quality of life of women with PD could be measured by various questionnaires such as Short-Form Health 12, Short-Form Health 36, or any other questionnaire used to evaluate quality of life. The clinical effective rate will be calculated from the total number of patients randomized. If the studies report the clinical effective rate exceeds 2 points of time, we will compare the records obtained at the baseline with the endpoint of the treatment.
3.2 Search strategy
We have customized search strategies with the help of an experienced librarian for each database. We will search the following sources without restrictions for date, language, or publication status: PubMed, Cochrane Central Register of Controlled Trials (CENTRAL) Cochrane Library, and EMBASE. We will apply a combination of Medical Subject Heading (MeSH) and free-text terms incorporating database-specific controlled vocabularies and text words to implement search strategies. Table 1 shows details of the search strategy for PubMed. We will also search the ongoing trial registered in the World Health Organization's International Clinical Trials Registry Platform. Besides, the previous relevant reviews conducted on complementary and alternative therapies for PD and reference lists of included studies will also be searched.
Table 1: Details of the search strategy for PubMed.
3.3 Identification and selection of studies
All studies identified through electronic and manual searches will be imported into NoteExpress literature management software (v3.2.0.7103) and the repeated literature will be removed electronically.
The process for identification and selection of studies will be conducted in 2 separate steps:
1. Two independent authors will screen each title and abstract of the literature and exclude the apparently unrelated research.
2. Literature preliminarily included will be acquired the full text for further inspection. Also 2 independent reviewers will assess eligibility of potentially relevant articles.
Conflicts will be resolved by discussion or seeking an independent third opinion.
3.4 Data extraction
Two independent authors will use a pretested and standardized Microsoft Excel data extraction form to extract the data from eligible studies. In case of disagreement, an independent third reviewer will be consulted. When there is missing information that may affect eligibility, we will try to contact the original study authors. We plan to extract the following data items from each eligible study:
Study characteristics : title, first author, year of publication, journal, registration number to trials registries, country of conduct, number of study arms, method of randomization and blinding, country of conduct, number of centers, and funding/sponsor.
3.4.1 Study participants’ characteristics
The characteristics of the study participants included gender, age, sociodemographic characteristics, sample size, number randomized to each arm, duration of disease, and disease diagnostic criteria.
3.4.2 Intervention characteristics
The intervention characteristics included type of therapy, frequency, the course of treatment, and therapy setting.
3.4.3 Outcome characteristics
The outcome characteristics included number of events, mean and standard errors, or standard deviation per arm.
3.5 Assessment of risk of bias
Risk of bias in the eligible studies will be assessed by the Cochrane Collaboration's risk of bias tool,[25]
3.6 Data analysis
3.6.1 Characteristics of the eligible studies
We will conduct descriptive statistics and describe the population characteristics of the eligible studies that include age, duration of disease, types of comparisons, duration of study, methodological variables and country, and so on.
3.6.2 Analytical procedure
We will perform a battery of pairwise meta-analyses to compare the active treatments that with the same interventions and the different control arms. The Bayesian network meta-analyses will be used to assess the relevant outcomes through comparing the different complementary and alternative therapies with control conditions from both direct and indirect comparisons.
3.6.3 Pairwise meta-analyses
We will perform pairwise meta-analyses for each pairwise comparison to obtain odds ratios for dichotomous outcomes or standardized mean differences for continuous outcomes, both with 95% credible intervals (CrIs). We will select a random effects model or the fixed effect model to synthesize data based on the value of I2 . Besides, we also use the value of I2 to measure the amount of heterogeneity.[26]
3.6.4 Network meta-analyses
We will perform a Bayesian network meta-analysis using the Markov chain Monte Carlo method in WinBUGS 1.4.3.[27] [28]
The ranking analysis of NMA can estimate the ranking probabilities of all treatments for each outcome. Then, we will use the surface under the cumulative ranking curve (SUCRA) values[29]
3.6.5 Assessment of inconsistency
When there are 3 treatments constituting a loop, we will evaluate the inconsistency among them. For the same comparison, we will estimate the agreement between direct and indirect effect. The node-splitting method will be used to evaluate consistency in the entire network using the mvmeta command in STATA.
3.6.6 Exploring heterogeneity, inconsistency, and sensitivity analyses
We expect small amounts of heterogeneity and inconsistency to be present considering the variety of study settings we plan to include. The following characteristics will be used to explore whether or not the treatment effects are robust for the 2 primary outcomes in subgroup analyses and network meta-regression:
1. Pain intensity at baseline
2. Study year
3. Study duration
4. Setting: individual vs group
5. Different types of patients, with different clinical outline concerning symptoms (if identified)
We will explore the sensitivity of our conclusions for the 2 primary outcomes, and the sensitivity will be analyzed by excluding the following aspects:
1. Studies with high risk of bias
2. Open studies
3. Studies providing published data only
4. Studies with unfair dose comparisons
3.6.7 Assessment of reporting bias
We will construct a comparison-adjusted funnel plots to assess the potential publication bias and small-study effects.[30]
3.6.8 Evaluating the quality of the evidence
We will evaluate the quality of the evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) instrument specifically developed for network meta-analysis.[31]
1. Present direct and indirect treatment estimates for each comparison of the evidence network.
2. Rate the quality of each direct and indirect effect estimate.
3. Present the NMA estimate for each comparison of the evidence network.
4. Rate the quality of each NMA effect estimate.
GRADE is characterized based on the study limitations, inconsistency, imprecision, indirectness, and publication bias. There are 4 levels classified by the GRADE for the quality of evidence: very low quality, low quality, moderate quality, and high quality.
4 Discussion
The evidence basis around the complementary and alternative therapies for PD has led to conflicting interpretation of results. Indeed, although several studies have assessed the comparative efficacy, the traditional pairwise meta-analyses only evaluated the direct comparison of pair-wised drugs and conflicting interpretation of results were also existed among different studies. This NMA aims to evaluate the current evidence for the efficacy and safety of complementary and alternative therapies for the PD. We will also get the treatment hierarchy of each competing treatment for the outcomes of interest. Its results will cause widespread concern: gynecologist, practice guideline developers, researchers, policymakers, as well as man whose wife or girlfriend suffered from dysmenorrhea.
Our study has several strengths. First, a comprehensive literature search will be conducted to include both published and unpublished work. Second, the GRADE will be used to rate the confidence in the estimates. Third, we will consider the potential sources of heterogeneity. Finally, a Bayesian framework will be used to pool the results and to estimate the ranking probabilities of all treatments for each outcome. On the other hand, some limitations are predictable. The publication bias exists for this review because of excluding non-English studies and some degree of clinical heterogeneity. We also hope that our findings will provide evidence and fill the gaps to reduce the uncertainty about the ranking of the interventions in terms of effectiveness and safety and encourage further research to enhance clinical decision-making.
Author contributions
Conceptualization: Fengting Zhai, Dongmei Wang.
Data curation: Fengting Zhai, Zhen Hua, Yuting Jiang.
Formal analysis: Fengting Zhai, Zhen Hua.
Funding acquisition: Dongmei Wang.
Methodology: Fengting Zhai, Zhen Hua, Yuting Jiang, Dandan Wang.
Resources: Dandan Wang.
Software: Fengting Zhai, Zhen Hua, Dandan Wang.
Supervision: Yuting Jiang.
Writing – original draft: Fengting Zhai.
Writing – review & editing: Fengting Zhai, Dongmei Wang.
References
[1]. Burnett M, Lemyre M. No. 345-primary dysmenorrhea consensus guideline. J Obstet Gynaecol Can 2017;39:585–95.
[2]. Rosenwaks Z, Seegar-Jones G. Menstrual pain: its origin and pathogenesis. J Reprod Med 1980;25:207–12.
[3]. Rencz F, Péntek M, Stalmeier PFM, et al. Bleeding out the quality-adjusted life years: evaluating the burden of primary dysmenorrhea using time trade-off and willingness-to-pay methods. Pain 2017;158:2259–67.
[4]. Proctor M, Farquhar C. Diagnosis and management of dysmenorrhoea. BMJ 2006;332:1134–8.
[5]. Unsal A, Ayranci U, Tozun M, et al. Prevalence of dysmenorrhea and its effect on quality of life among a group of female university students. Ups J Med Sci 2010;115:138–45.
[6]. Banikarim C, Chacko MR, Kelder SH. Prevalence and impact of dysmenorrhea on Hispanic female adolescents. Arch Pediatr Adolesc Med 2000;154:1226–9.
[7]. László KD, Gyrffy Z, Ádám S, et al. Work-related stress factors and menstrual pain: a nationwide representative survey. J Psychosom Obstet Gynaecol 2008;29:133–8.
[8]. Burnett MA, Antao V, Black A, et al. Prevalence of primary dysmenorrhea in Canada. J Obstet Gynaecol Can 2005;27:765–70.
[9]. Sundell G, Milsom I, Andersch B. Factors influencing the prevalence and severity of dysmenorrhoea in young women. Br J Obstet Gynaecol 1990;97:588–94.
[10]. Dawood MY. Nonsteroidal anti-inflammatory drugs and changing attitudes toward dysmenorrhea. Am J Med 1988;84:23–9.
[11]. Hofmeyr GJ. Bassin J. Dysmenorrhoea. Topics in Obstetrics and Gynaecology. Johannesburg: Julmar Communications; 1996. 269–74.
[12]. Jones AE. Managing the pain of primary and secondary dysmenorrhoea. Nurs Times 2004;100:40–3.
[13]. Marjoribanks J, Ayeleke RO, Farquhar C, et al. Nonsteroidal anti-inflammatory drugs for dysmenorrhoea. Cochrane Database Syst Rev 2015;(7):CD001751.
[14]. Tziona P, Theodosis-Nobelos P, Rekka EA. Medicinal chemistry approaches of controlling gastrointestinal side effects of non-steroidal anti-inflammatory drugs. Endogenous protective mechanisms and drug design. Med Chem 2017;13:408–20.
[15]. Abaraogu UO, Igwe SE, Tabansi-Ochiogu CS. Effectiveness of SP6 (Sanyinjiao) acupressure for relief of primary dysmenorrhea symptoms: a systematic review with meta and sensitivity analyses. Complement Ther Clin Pract 2016;25:92–105.
[16]. Shi S, Klotz U. Clinical use and pharmacological properties of selective COX-2 inhibitors. Eur J Clin Pharmacol 2008;64:233–52.
[17]. Uysal G, Akkaya H, Cagli F, et al. A comparison of two different oral contraceptives in patients with severe primary dysmenorrhoea. J Obstet Gynaecol 2018;38:828–32.
[18]. Iwasaki Y, Morita A, Iwasawa T, et al. A nonpungent component of steamed ginger-[10]-shogaol-increases adrenaline secretion via the activation of TRPV1. Nutr Neurosci 2006;9:169–78.
[19]. Campbell MA, McGrath PJ. Non-pharmacologic strategies used by adolescents for the management of menstrual discomfort. Clin J Pain 1999;15:313–20.
[20]. Navvabi Rigi S, Kermansaravi F, Navidian A, et al. Comparing the analgesic effect of heat patch containing iron chip and ibuprofen for primary dysmenorrhea: a randomized controlled trial. BMC Womens Health 2012;12:25.
[21]. Chaudhuri A, Singh A, Dhaliwal L. A randomised controlled trial of exercise and hot water bottle in the management of dysmenorrhoea in school girls of Chandigarh, India. Indian J Physiol Pharmacol 2013;57:114–22.
[22]. Chung YC, Chen HH, Yeh ML. Acupoint stimulation intervention for people with primary dysmenorrhea: systematic review and meta-analysis of randomized trials. Complement Ther Med 2012;20:353–63.
[23]. Shamseer L, Moher D, Clarke M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;350:g7647.
[24]. Hutton B, Salanti G, Caldwell DM, et al. The PRISMA extension statement for reporting of systematic reviews incorporating network meta-analyses of health care interventions: checklist and explanations. Ann Intern Med 2015;162:777–84.
[25]. Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration 2011;
http://handbook.cochrane.org/ [26]. Higgins JPT, Altman DG. Deeks JJ, Higgins JPT. Chapter 9: Analysing data and undertaking meta-analyses. John Wiley & Sons, Cochrane Handbook for Systematic Reviews of Interventions.. Chichester (UK): 2011.
[27]. Lunn DJ, Thomas A, Best N, et al. WinBUGS – a Bayesian modeling frame work: concepts, structure, and extensibility. Statist Comput 2000;10:325–37.
[28]. White IR, Barrett JK, Jackson D, et al. Consistency and inconsistency in network meta-analysis: model estimation using multivariate metaregression. Res Synth Methods 2012;3:111–25.
[29]. Salanti G, Ades AE, Ioannidis JP. Graphical methods and numerical summaries for presenting results from multiple-treatment meta-analysis: an overview and tutorial. J Clin Epidemiol 2011;64:163–71.
[30]. Egger M, Smith GD, Schneider M, et al. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34.
[31]. Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011;64:383–94.
