This protocol has been registered in the PROSPERO database with ID CRD42021265201, following the PRISMA criteria under the following research question:

P: Adult patients with temporomandibular internal disorders with an arthroscopy indication.

I: Temporomandibular joint arthroscopy followed by intra-articular injections.

C: Operative or diagnostic arthroscopy with intra-articular infiltrations of different substances.

O: Impact on the clinical conditions of the patient is evaluated by variables such as pain and maximum mouth opening.

Focused question: Which of the substances usually injected intra-articularly after arthroscopy shows better clinical results, such as pain relief and maximal mouth opening (MMO)?

Criteria for Selecting Studies

Inclusion criteria The inclusion criteria were articles that reported infiltration of different substances only after arthroscopy procedures in patients over 15 years, describing visual analog scales (VAS) and (MMO), including clinical cases, case series, observational studies, and randomized clinical trials with at least 3 months of follow-up. The selection process began with articles published in the English language between January 1989 and December 2020.

Exclusion criteria Studies that included arthrocentesis, animal studies, connective tissue disease, and articles that included patients with previous surgical treatment.

Types of substances Hyaluronic acid (HA) and sodium hyaluronate (SH), platelet-rich plasma (PRP), plasma rich in growth factors (PRGF), corticoids, and analgesics

Search Strategy

A systematic review was performed according to the PRISMA statement, and the protocol was enrolled and recorded in PROSPERO-CRD42021265201. A search of the Web of Science, MEDLINE, and EMBASE databases was done. The MEDLINE searches included a combination of relevant search terms. The search was completed on 21/09/2022. The results were limited to human-subject, and English-language articles. All abstracts were analyzed, and full-text articles were obtained when inclusion criteria were fulfilled. The references of the subsequent full-text articles were reviewed to identify additional relevant articles.

Search Methods for the Identification of Studies

A generic search strategy composed of controlled vocabulary exploded as Mesh (Medical Subject Headings) and free language, considering synonyms, abbreviations, acronyms, spelling, and plural variations, was designed. Individual search strategies were developed for each source of information (Appendix 1).

Critical Appraisal and Assessment of the Risk of Bias in the Included Studies

All studies were evaluated independently and duplicated by two reviewers (JPL and MPO) to determine methodological quality using the ROB2 tool (Version 2 of the Cochrane risk-of-bias tool for randomized trials) for randomized clinical trials. This tool is based on the evaluation and qualification of clinical studies, considering five domains, and focuses on assessing aspects that are relevant to the risk of bias in a study of this type (trial design, conduct, and reporting). For each domain, judgment can be ‘Low’ or ‘High’ risk of bias, or it can express ‘Some concerns’, and likewise, in each domain, there is a space for the evaluator to provide his or her personal opinion about it. Finally, any disagreements among the reviewers were subjected to the evaluation of a third evaluator (LVG).

Data Collection Process

The list with the bibliographic references identified in the electronic searches was downloaded into a library of the Rayyan® program, where duplicate publications were eliminated, and an initial screening was carried out. In the first instance, the reviewers identified eligible articles by title. Afterward, each of the authors separately assessed the abstracts of these articles and selected potentially eligible studies. The reviewers subsequently independently verified the eligibility criteria (inclusion and exclusion) by reviewing each full-text publication (Table 1).

Table 1 Summary of demographic information of all studies includedData Extraction

The characteristics of the selected evidence were summarized according to what was reported in the original publications using a standardized data extraction format in chronological order. The data collected included author, year, study design, population, type of substance used (HA/SH, PRP, PRGF, corticoids, and analgesics), preoperative and postoperative VAS, preoperative and postoperative MMO, and follow-up period. In studies where the data corresponding to the summary and dispersion measures were not precisely specified, their graphical representations were used to extract the data using a plotdigitizer (https://plotdigitizer.com/app).

Data Synthesis

For the main comparisons between the results of the different combination techniques, see Table 2, accompanied by a narrative synthesis. Subsequently, two random effects meta-analyzes were performed to compare weighted means between the pre- and postoperative periods of the VAS and MMO variables using the RevMan 5.4.1 software developed by the Cochrane Collaboration. Each meta-analysis considered two subgroups, one to estimate the effect of PRGF and HA; the other groups were not considered due to the heterogeneity between presentations or the lack of data availability. A difference in measures (MD) with a confidence interval (CI) excluding 0 was considered statistically significant.

Table 2 Summary of outcomes