Introduction

Lower third molar extractions may be associated with sequelae, such as pain, edema, mouth-opening difficulties, and alveolar osteitis, harming patients’ postoperative quality of life^1,2. Several therapeutic approaches have been indicated to control these inflammatory signs and symptoms, including drug therapy, cryotherapy, low-level laser therapy, and blood concentrates^2–7.

Blood concentrates originate from autologous blood and primarily consist of platelets, fibrin, leukocytes, and growth factors⁸. Platelet-rich plasma (PRP) is the first-generation blood concentrate, presenting disadvantages such as the need for more than one step for production and anti- and procoagulant addition^9,10. Studies have developed blood products, such as platelet-rich fibrin (PRF), later renamed leukocyte- and platelet-rich fibrin (L-PRF)¹¹. New centrifugation protocols developed from L-PRF resulted in advanced platelet-rich fibrin (A-PRF), advanced platelet-rich fibrin plus (A-PRF+), and concentrated growth factor (CGF)^10–12.

Blood concentrates are fibrin-based biomaterials containing cells, matrix proteins, pro- and anti-inflammatory mediators, and growth factors, showing positive performance in handling postoperative sequelae, such as pain, edema, trismus, and alveolar osteitis^13,14. Hence, inflammatory response modulation occurs, optimizing primary hemostasis, chemotaxis, angiogenesis, and mitogenesis of endothelial cells¹⁴.

Systematic reviews have analyzed the effect of blood concentrates on managing postoperative sequelae, such as pain, edema, mouth opening, and alveolar osteitis, showing conflicting results and several recommendations for randomized clinical trials with larger samples and more homogeneous methodologies^15–18. Moreover, the overall syntheses and assessments of these reviews have not occurred, which is relevant considering each review answers different research questions on the same subject. These reviews analyze different primary studies and blood concentrates, causing the referred outcome variability. Therefore, this overview summarized the available evidence and answered the following question: What is the current knowledge on the performance of blood concentrates in handling sequelae after lower third molar extractions with the evidence available in systematic reviews?

Methods

Protocol registration

The protocol of this review was created according to the PRISMA-P¹⁹ guidelines and registered a priori in the PROSPERO database (CRD42023405301) (https://www.crd.york.ac.uk/prospero/). The overview was reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations²⁰.

Eligibility criteria

The study included systematic reviews with or without meta-analyses investigating the performance of blood concentrates in controlling sequelae after lower third molar extractions. The four outcomes analyzed were pain, edema, mouth opening, and alveolar osteitis. There were no restrictions on year or language.

The exclusion criteria were:

• Randomized clinical trials;

• Studies with different objectives;

• Case reports;

• Animal studies;

• In-vitro studies;

• Letters to the editor;

• Observational studies;

• Systematic reviews with or without meta-analyses and with outcomes not directly related to pain, edema, mouth opening, and alveolar osteitis.

Sources of information, search, and selection of studies

The electronic searches were performed in June 2022 and updated in May 2024 in the Cochrane Library, Embase, Latin American and Caribbean Health Sciences Literature (LILACS), MEDLINE (via PubMed), Scientific Electronic Library Online (SciELO) databases, and Scopus and Web of Science citation databases. The EASY and the MedRxiv preprint databases partially captured the grey literature. A search strategy was formulated for each database, respecting their syntax rules and using the Health Science Descriptors (DeCS), Medical Subject Headings (MeSH), and Embase Subject Headings (Emtree) descriptors with the Boolean operators AND/OR to maximize the search. Table 1 describes the search strategy for each database.

The search results were exported to EndNote Web^TM software (Thomson Reuters, Toronto, ON, Canada) to identify and remove duplicates. Next, they were exported to Rayyan QCRI software (Qatar Computing Research Institute, Doha, Qatar) for title and abstract analyses according to the described eligibility criteria. Subsequently, the full texts of the preliminary eligible studies were obtained and evaluated. Microsoft Word 2010 (Microsoft, Redmond, WA, United States of America) assisted in simultaneously evaluating grey literature results. Two eligibility reviewers (VLA and CMM) performed this entire process independently. Divergences were solved after consulting with a third eligibility reviewer (LRP).

Data collection

After selecting the articles, two reviewers (VLA and CMM) extracted the following information: author, year, and journal of publications; research objective; blood concentrate type; searched databases and period; primary studies included in the systematic reviews; risk of bias assessment tools of the studies included in the systematic reviews; and primary results and conclusions. The reviewers (VLA and CMM) underwent a calibration exercise to ensure consistency during data extraction by jointly extracting information from an eligible study. In case of missing data, the corresponding authors of the eligible studies were contacted via e-mail, with three weekly contact attempts for up to a month.

Critical assessment of individual evidence sources

Three reviewers (VLA, CMM, and LRP) analyzed the methodological quality of the included systematic reviews using the AMSTAR-2 critical assessment tool²¹. This tool contains 16 questions (seven critical and nine non-critical), and their answer options were “yes,” “partially yes,” “no,” or “no meta-analysis performed.” The methodological quality was low if one critical question had a “no” response and critically low if two or more critical questions had “no” answers.

Risk of individual bias in the studies

Three reviewers (VLA, CMM, and LRP) independently assessed the risk of bias in the included systematic reviews using the ROBIS tool for evaluating the risk of bias in systematic reviews²².

Initially, they analyzed four domains from which biases might have been introduced in a systematic review:

• Eligibility criteria;

• Identification and selection of articles;

• Data collection and study assessment;

• Synthesis and results.

Moreover, each domain has five or six signaling questions with the following possible answers: “yes (Y),” “probably yes (PY),” “probably no (PN),” “no (N),” “not informed (NI),” or “not applicable (NA).” Later, the overall risk of bias was evaluated during the interpretation of review findings, considering whether it identified limitations within the described domains.

All reviewers agreed on the decisions about the scoring system and cutoffs before analyzing biases. The authors determined the classification system for the bias of each domain as a “low risk” if all signaling questions scored Y/PY, “unclear risk” if only one question received PN/N/NI, and “high risk” if more than one question scored PN/N/NI. Moreover, the overall risk of bias regarding each systematic review was low if all four domains had a “low risk” or only one had an “unclear risk,” moderate if two or more domains had an “unclear risk,” and high if one or more domains had a “high risk.”

Results

Selection of systematic reviews

The study identification process found 690 records. After removing duplicates and reading the titles and abstracts, only 17 records were fully assessed for eligibility criteria, of which two were eliminated. Finally, 15 systematic reviews were selected^{15–18,23–33}. The list of references of the eligible studies were carefully searched but did not reveal additional systematic reviews. Figure 1 details the study selection in a flowchart.

Characteristics of the included systematic reviews

The analysis comprised systematic reviews published between 2007 and 2024. One study was published in Chinese²⁸, and the others in English. There were similarities in the search databases, with all reviews including MEDLINE (via PubMed). Additionally, seven systematic reviews manually searched journals or the list of references of their primary studies^{16,17,23,24,29,31,32}. Table 2 describes the main characteristics of the included systematic reviews.

These reviews evaluated 75 primary studies published between 2007 and 2023, with 32 common studies across multiple systematic reviews. Figure 2 illustrates the overlap of primary studies.

Considering the heterogeneity level regarding primary studies’ designs (parallel or split mouth), samples, measurement methods, and differences in the postoperative assessment of pain, edema, and trismus, two systematic reviews could not perform a meta-analysis for these outcomes^15,33. Table 3 describes the main results of the systematic reviews.

Critical assessment of individual evidence sources

According to the AMSTAR-2²¹ analysis, only one study presented high methodological quality³³, while six studies^{15,16,18,25,31,33} demonstrated low methodological quality, and the other systematic reviews were critically low^{17,23,24,26–30}. Table 4 summarizes the responses of each systematic review to the AMSTAR-2²¹ questionnaire.

Risk of individual bias of the eligible studies

Only two systematic reviews^31,32 showed low risk, and the others had a high risk of bias^{15–18,23–30,33}. There were relevant concerns regarding the risk of bias, including:

• The absence of a previous study protocol registration;

• The absence of a comprehensive electronic search in grey literature databases and with additional strategies;

• The absence of a definite mention of at least two eligibility reviewers in the methodological steps of the systematic review;

• The absence of sensitivity analyses;

• Primary studies with a high risk of bias;

• Protocol deviations;

• Data interpretation focused on statistically different results.

Figure 3 summarizes the overall findings of the risk of bias assessment.

Discussion

The present overview analyzed the current knowledge on the performance of blood concentrates in handling sequelae after lower third molar extractions with the evidence available in systematic reviews. Considering the surgical nature and complexity, third molar removal may be associated with postoperative sequelae, such as pain, edema, mouth-opening difficulties, and alveolar osteitis². The healing process involves a highly coordinated sequence of biochemical, physiological, cellular, and molecular responses to restore tissue integrity. Blood concentrates have been extensively used to improve tissue repair and modulate the postoperative inflammatory process^34,35.

Only two systematic reviews^15,32 evaluated PRP, and the evidence available for using PRP in lower third molar extractions is scarce. Barona-Dorado et al.¹⁵ included three primary studies^36–38 in the cited systematic review¹⁵, showing conflicting results regarding the efficacy of PRP in controlling pain, mouth opening, and alveolar osteitis. Two primary studies^37,38 did not present methodological flaws in the clinical trials but showed result interpretation biases. One of the articles³⁸ affirmed that PRP significantly reduced pain levels but did not provide any value (mean and standard deviation). It did not determine the method for diagnosing alveolar osteitis and analyzed the variables jointly between upper and lower third molar sockets either. The other study³⁷ did not provide objective data for the analyzed variables and had a small sample. It is worth noting that this included systematic review¹⁵ also presents methodological performance biases. The risk of bias analyzed in the ROBIS tool²² showed that, despite determining the study objective, there was no review question. Moreover, the review neither performed a manual search in the references of the included primary studies nor mentioned the participation of two reviewers during data collection and the risk of bias classification in said studies.

Costa et al.³² selected nine studies assessing PRP performance^39–47. Five of these articles^{39,40,43,45,46} reported statistically favorable outcomes for the group treated with the blood concentrate, four^40,42,43,45 concluded that PRP significantly reduced edema compared to blood clots, and three^43,45,46 presented a lower occurrence of trismus. Although the study by Costa et al.³² had a high methodological quality according to AMSTAR 2²¹ and a low risk of bias according to ROBIS²², the results require careful interpretation, as the primary studies assessing PRP performance exhibited a moderate to high risk of bias.

Conflicting findings also appeared for the role of L-PRF in controlling postoperative sequelae after lower third molar removal. Most included systematic reviews presented favorable conclusions regarding the efficacy of blood concentrates^{16,17,24,26–30,32}. However, it is worth noting that three of these investigations^16,27,28,32 also highlighted outcome limitations and the need for randomized clinical trials with larger samples and a more homogeneous methodology for a higher comparison effect. Except for the study by Costa et al.³², the risk of bias analysis with the ROBIS tool²² revealed that all reviews present at least a few methodological flaws concerning the lack of reference to a registration protocol^24,26–29, language limitations^17,26,30, or the absence of a definite mention of at least two eligibility reviewers during study selection^24,28, data extraction¹⁶, and risk of bias assessment^24,26,28 of primary studies. Some methodological flaws, such as the lack of a previous registration protocol, may promote imprecise conclusions, as it is hard to know whether these criteria were established in advance and guided the reviewers’ steps during the review or were determined or modified in the review process^21,22. Therefore, the findings of the referred systematic reviews must be carefully interpreted considering the identified methodological limitations to prevent precipitated or imprecise conclusions.

Some systematic reviews included in the present overview did not have sufficient evidence to indicate the use of PRF after lower third molar extraction2^3,25,31. Including patients with lower third molars at varying degrees of inclusion/impaction in a single sample may have introduced biases in clinical trial results²³. Additionally, surgical difficulty presents a statistically significant relationship with the intensity of postoperative sequelae^48,49. It is worth noting that one systematic review²³ presented a high risk of methodological bias due to the lack of reference to previous study protocol registration, the absence of a comprehensive electronic search in grey literature databases, the absence of a definite mention of at least two eligibility reviewers during study selection and risk of bias assessment, and the absence of sensitivity analyses. Another systematic review with a high risk of bias²⁵ needs careful interpretation because, except for one of its primary studies⁵⁰, the others did not present randomized clinical trial registrations, and the statistically insignificant results might have been hidden in the final publication, introducing result interpretation biases. Also, the primary studies of one of the included systematic reviews³¹ were split mouth and double-blind clinical trials, and the statistical methods used in the meta-analysis allowed a reliable interpretation of results. The statistical methods of the meta-analysis allowed a reliable interpretation of results, but the methodological heterogeneity in measuring edema and insufficient information on alveolar osteitis limited evidence availability. Consequently, a meta-analysis for postoperative pain was only feasible in three randomized clinical trials from the same systematic review³¹, highlighting the limitations of available evidence.

Our overview showed that A-PRF reduced pain in the first^30,32, second¹⁸, third^18,27,32, and seventh³² postoperative days. As for trismus and edema, A-PRF promoted higher reduction during and after the third postoperative day^18,32. These results seem debatable because the release of growth factors through A-PRF occurs for 10 days, and the physiological inflammatory response appears immediately after the tissue lesion^51,52. Thus, these data should be carefully interpreted because randomized clinical trials presented methodological heterogeneity, potential result interpretation biases18, methodological flaws for not mentioning the predefined registration protocol²⁸, restrictions on publication language18, or the absence of two reviewers during study selection and data extraction of the primary studies²⁸.

While systematic reviews are essential for evidence-based decision-making, accepting their results without a critical assessment may provide improper conclusions²¹. A critical evaluation should involve deep knowledge of the referred therapy and the physical and biological blood concentrate properties, indications, advantages, and limitations. Therefore, the limitations of this overview stem from the high heterogeneity and risk of bias of the included systematic reviews and their primary studies, complicating the establishment of definitive evidence. However, it is worth noting that the data for clinical decision-making on blood concentrate use should be carefully interpreted, as a detailed analysis showed relevant methodological flaws in apparently robust systematic reviews on the topic.

Additionally, systematic reviews of high-quality randomized controlled trials are essential to provide accurate evidence on this topic. In this aspect, high-quality randomized controlled trial should present an adequate randomization process, sample size calculation, blinding of operator, patient and outcome assessor, an adequate follow-up period, and protocol registration.

Conclusion

Biological factors inherent to blood concentrates are promising for optimizing tissue repair. However, the current evidence from systematic reviews remains regarding the efficacy of these products in managing sequelae after lower third molar extractions.

Randomized clinical trials with methodological standardization are first required for considering systematic reviews as adequate references for evidence-based decision-making. Moreover, systematic reviews must be performed according to scientifically established guidelines to minimize biases and ensure methodological rigor.

Acknowledgements

Not applicable.

Data availability statement

All data sets were generated or analyzed in the current study.

References

Edited by

Publication Dates

History