This protocol was developed according to the recommendations of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol guidelines (PRISMA-P) (27).
The protocol is registered with the International Prospective Register of Systematic Reviews. (PROSPERO CRD 42020163438)
A comprehensive search strategy with the help of an expert librarian will be devised. Keywords, texts words and synonymous primarily conceived from the PICO* framework will be searched from MeSH* database, previous research articles, and books. PubMed, EMBASE, Scopus and Cochrane library databases will be utilized to searach relevant studies. There will be no time or language limitation. An advanced search strategy on PubMed is conducted for exploring relevant studies as an example. (Annexure-1)
Grey literature and unpublished studies will be searched through Google Scholar and ProQuest.
Reference list of identified studies will be manually checked for identification of further studies. Three endodontic journals: Journal of Endodontics, International Endodontic Journal and, Australian Endodontic Journal will be manually searched from 2000-2019.
All selected studies from various sources will be transported to EndNote Software X9. One reviewer will make an initial screening of studies by titles and abstracts to remove irrelevant and duplicate studies. The full text of the remaining studies will then be assessed by the two reviewers (DS, ZA) using the inclusion and exclusion criteria mentioned above. If there will be any disagreement over the eligibility of any study, it will be resolved through discussion. The whole process of studies selection will be done in a transparent, systematic and reproducible way and will be presented in the PRISMA flow diagram (27).
A standardized data extraction form on an Excel sheet will be created with a consensus of two reviewers (DS, ZA). The data will include study ID, sample size in the intervention and the control groups, diagnosis, type of teeth, concentration of irrigant, characteristics of intervention groups and control groups, preoperative pain, postoperative pain on 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, days and conclusion. Data extraction will be done by the two reviewers (DS, ZA) independently. Before data extraction, the two reviewers will be calibrated in order to achieve consistency in data extraction. Authors of included studies will be contact where necessary information will be needed.
Risk of Bias Assessment
Quality assessment of studies will be carried out as a part of data extraction using the Cochrane Risk of Bias Assessment Tool. We will evaluate six domains in each study under this tool: Random sequence generation, Allocation concealment, Blinding of participants and operators, Blinding of outcome assessors, incomplete outcome data, and other methodologic bias. The study will be ranked as low risk of bias if all domains are adequately reported in the text, High risk of bias if all or anyone domain is not reported in the text, and Unclear risk of bias if any one of the domain is not described satisfactorily in the text.
Strategies for data synthesis
We will synthesize the data both quantitatively and qualitatively. In qualitative analysis, various characteristics of included studies will be evaluated and compared. In addition, we will estimate the pooled effect size of postoperative pain in the form of weighted mean difference (WMD) or Standardized mean difference (SMD) (in case of different scales of pain measurements are used). Pooled effect size will be displayed in the form of the forest plot. We will used RevMan 5.3 for statistical analysis of the data.
We are expecting heterogeneity among studies due to variability in study designs, clinical methodology, different types of teeth and variable pulpal and periapical status of teeth, and sampling variability. Heterogeneity among studies will be estimated by I2 statistics and P-value. The degree of heterogeneity will be considered acceptable if I2 <30%, Moderate heterogeneity if I2 is 30-60%, and substantial heterogeneity if I2> 60%.
Due to the anticipated heterogeneity among study, random-effect model will be used for meta-analysis. A random-effect model allows variability in study designs, methodology and especially sampling variability. Sensitivity analysis will also be conducted in order to evaluate the robustness of results from the meta-analysis.
Subgroup analysis & Publication bias.
If we found a substantial heterogeneity, the reasons behind the heterogeneity will be explored. All possible factors responsible for heterogeneity may be further divided into subgroups. Subgroup analysis will be performed to evaluate heterogeneity. Publication bias within studies will be assessed by funnel plot.