Systematic Review

Corticosteroids for managing acute pain

subsequent to surgical extraction of
mandibular third molars
A systematic review and meta-analysis
Anna Miroshnychenko, MSc; Maria Azab, BHSc; Sara Ibrahim, BHSc; Yetiani Roldan, MD;
Juan Pablo Diaz Martinez, MSc; Divyalakshmi Tamilselvan, BHSc; Leon He, BHSc;
Olivia Urquhart, MPH; Francisca Verdugo-Paiva, DDS, MSc; Malavika Tampi, MPH;
Deborah E. Polk, PhD; Paul A. Moore, DMD, PhD, MPH; Elliot V. Hersh, DMD, MS, PhD;
Romina Brignardello-Petersen, DDS, MSc, PhD; Alonso Carrasco-Labra, DDS, MSc, PhD

ABSTRACT

Background. Corticosteroids are used to manage pain after surgical tooth extractions. The authors
assessed the effect of corticosteroids on acute postoperative pain in patients undergoing surgical
tooth extractions of mandibular third molars.
Types of Studies Reviewed. The authors conducted a systematic review and meta-analysis. The
authors searched the Epistemonikos database, including MEDLINE, Embase, Cochrane Central
Register of Controlled Trials, and the US clinical trials registry (ClinicalTrials.gov) from inception
until April 2023. Pairs of reviewers independently screened titles and abstracts, then full texts of
trials were identified as potentially eligible. After duplicate data abstraction, the authors conducted
random-effects meta-analyses. Risk of bias was assessed using Version 2 of the Cochrane Risk of Bias
tool and certainty of the evidence was determined using the Grading of Recommendations
Assessment, Development and Evaluation approach.
Results. Forty randomized controlled trials proved eligible. The evidence suggested that cortico-
steroids compared with a placebo provided a trivial reduction in pain intensity measured 6 hours
(mean difference, 8.79 points lower; 95% CI, 14.8 to 2.77 points lower; low certainty) and 24 hours
after surgical tooth extraction (mean difference, 8.89 points lower; 95% CI, 10.71 to 7.06 points
lower; very low certainty). The authors found no important difference between corticosteroids and a
placebo with regard to incidence of postoperative infection (risk difference, 0%; 95% CI, –1% to 1%;
low certainty) and alveolar osteitis (risk difference, 0%; 95% CI, –3% to 4%; very low certainty).
Practical Implications. Low and very low certainty evidence suggests that there is a trivial dif-
ference regarding postoperative pain intensity and adverse effects of corticosteroids administered
orally, submucosally, or intramuscularly compared with a placebo in patients undergoing third-
molar extractions.
Key Words. Corticosteroids; surgical tooth extraction; third molars; postoperative pain; evidence-
based dentistry.
JADA 2023:154(8):727-741
https://doi.org/10.1016/j.adaj.2023.04.018

S
urgical removal of impacted mandibular third molars is 1 of the most frequently performed
surgical interventions in dental surgery, with more than 10 million teeth extracted per year.1
Copyright ª 2023
The most common complications, including pain, swelling, and trismus, can severely affect a American Dental
patient’s quality of life during the immediate postoperative period. Analgesics and anti- Association. This is an
inflammatory drugs prescribed postoperatively should relieve pain, reduce swelling and trismus, open access article under
the CC BY-NC-ND license
and improve healing without undesirable adverse effects. Therefore, medications that exert both
(http://creativecommons.
analgesic and anti-inflammatory effects, such as corticosteroids, could be used for the management org/licenses/by-nc-nd/
of postoperative discomfort. 4.0/).

JADA 154(8) n http://jada.ada.org n August 2023 727

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Corticosteroids can be divided into 2 major groups, that is, glucocorticoids and mineralocorti-
coids. Glucocorticoids are used for the management of postoperative complications after surgical
tooth extraction because of their substantial anti-inflammatory effects.2 The term corticosteroids
will be used to represent glucocorticoids in our review.
Corticosteroids are classified according to their duration of action and relative anti-inflammatory
potency compared with hydrocortisone, a reference standard with a potency of 1. The higher the
relative potency anti-inflammatory score, the higher the corticosteroid’s anti-inflammatory potency.
Short-acting glucocorticoids include cortisol and cortisone, with a duration of action of fewer than 12
hours and anti-inflammatory potency of 1.3 Intermediate-acting corticosteroids include prednisone and
prednisolone, with an anti-inflammatory potency of 4, and 6-methylprednisolone and triamcinolone,
which both have an anti-inflammatory potency of 5. Intermediate corticosteroids have a duration of
action of 12 through 36 hours. The long-acting glucocorticoids include dexamethasone, with a
duration of action of more than 36 hours and anti-inflammatory potency of 25.3 The administration of
corticosteroids in dentistry typically varies among oral, intramuscular, and submucosal routes.
Available systematic reviews (SRs) to inform the effect of corticosteroids for managing postoperative
acute pain consecutive to surgical tooth extractions have several limitations. Almost all were published
before 2018,4-7 with only 1 published in 2020.8 They did not use the Grading of Recommendations
Assessment, Development and Evaluation (GRADE)9 approach to assess the certainty of the evi-
dence—an essential component of SRs and a key step to inform the formulation of guideline rec-
ommendations—and they applied suboptimal methods to synthesize pain-related outcomes.
The purpose of our SR was to determine the effect of corticosteroids administered orally, sub-
mucosally, or intramuscularly on the management of pain subsequent to surgical tooth extraction,
including impacted mandibular third-molar extractions. Our review informed the clinical questions
posed by the forthcoming evidence-based clinical practice guideline for the pharmacologic man-
agement of acute dental pain consecutive to tooth extractions (A. Carrasco-Labra D.E. Polk, O.
Urquhart, and colleagues, unpublished data, 2023). This review and associated clinical practice
guidelines were led by the American Dental Association Science and Research Institute, the School
of Dental Medicine at the University of Pittsburgh, the Center for Integrative Global Oral Health at
the University of Pennsylvania, and a guideline panel including primary care dentists, oral and
maxillofacial surgeons, public health practitioners, pharmacoepidemiologists, biostatisticians, and
health research methodologists, among others.

METHODS
This article follows the guidelines of the Preferred Reporting Items for Systematic Reviews and
Meta-Analyses statement10 (eTable 1, available online at the end of this article). We also followed
established methodological considerations defined for evidence synthesis and guideline develop-
ment, used eligibility criteria determined per the recommendation questions that the guideline
panel proposed, and used information outlined in the National Academy of Medicine’s Framing
Opioid Prescribing Guidelines for Acute Pain: Developing the Evidence.11

Eligibility criteria
We included randomized controlled trials (RCTs) that compared the effect of corticosteroids
administered orally, submucosally, or intramuscularly at any dose with that of a placebo in
adolescent, adult, or older adult participants undergoing surgical (that is, extraction of a tooth
with the need of a flap and osteotomy) third-molar extraction, regardless of the language of pub-
ABBREVIATION KEY lication. We included the following continuous outcomes: pain intensity at 6 hours, pain at
GI: Gastrointestinal. 24 hours, total pain relief at 6 hours, and global efficacy rating at 6 hours. We included the following
GRADE: Grading of dichotomous outcome: adverse effects (for example, postoperative surgical site infection, alveolar
Recommendations osteitis, mood alteration, and gastrointestinal [GI] adverse effect at any time). We excluded studies
Assessment,
Development and
that administered corticosteroids intravenously and studies that only reported outcomes associated
Evaluation. with the management of inflammatory complications (that is, trismus, facial swelling, or infection).
L$OVE: Living Overview of
Evidence. Search methods to identify and select studies
NR: Not reported.
We performed the evidence search in 2 steps. First, we conducted searches in the Epistemonikos
RCT: Randomized
controlled trial. database, a comprehensive mate-search engine and updated source of relevant SRs and primary studies
SR: Systematic review. to inform health decision making (Appendix, available online at the end of this article). Using

728 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
artificial intelligence technology, Epistemonikos periodically screens across the following databases:
Cochrane Database of Systematic Reviews, PubMed, MEDLINE, Embase, Cumulative Index to
Nursing and Allied Health Literature, PsycINFO, Latin American and Caribbean Health Sciences
Literature, Database of Abstracts of Reviews of Effects, Health Technology Assessment Database,
Campbell Collection online library, Joanna Briggs Institute Database of Systematic Reviews and
Implementation Reports, Evidence for Policy and Practice Centre Evidence Library, and the US
clinical trials registry (ClinicalTrials.gov).12 Our search strategy included a combination of free and
controlled key words. These terms include specific key words to represent the concepts of surgical
extraction, third molars, and corticosteroids (Appendix, available online at the end of this article).
In addition, we used the Living Overview of Evidence (L$OVE) platform (Epistemonikos
Foundation), which maps the question of interest to a repository maintained through additional
searches on PubMed, Embase, and Cochrane Central Register of Controlled Trials. We searched
L$OVE for third-molar–related literature without restriction according to study design, language, or
publication status.13 The search covered the period from the inception date of each database
through April 2023 and had no language restrictions. The results of the searches in each source were
deduplicated by means of an algorithm that compares unique identifiers (that is, database identi-
fication, digital object identifier, and trial registry identification) and citation details (that is, author
names, journal, year of publication, volume, number, pages, article title, and article abstract).
Pairs of reviewers (A.M., M.A., Y.R., J.P.D.M., D.T., L.H., F.V.-P.) independently evaluated the
titles, abstracts, and full text of potentially eligible studies across all databases. When eligibility
consensus was elusive, a third reviewer served as arbiter (R.B.-P., A.C.-L.).

Data collection
After training and calibration exercises, pairs of reviewers independently extracted data for each
eligible trial using a standardized, pilot-tested data extraction form. We collected information on
trial characteristics (for example, intervention, comparison, and co-interventions), patient char-
acteristics (for example, age, sex, country, and type of extraction), and outcomes of interest. Re-
viewers resolved discrepancies by means of discussion, and, when necessary, a third reviewer served
as arbiter.

Risk of bias in studies

After training and calibration exercises, pairs of reviewers used Version 2 of the Cochrane Risk of
Bias tool for each eligible trial and outcome to assess risk of bias in RCTs, rating trials as being at
low risk of bias, probably at low risk of bias, probably at high risk of bias, or at high risk of bias across
the domains of bias arising from the randomization process, bias due to deviations from the intended
intervention, bias due to missing data, bias due to measurement of the outcome, and bias in se-
lection of the reported results.14 We rated trials as high risk of bias overall if 1 or more domains were
rated as probably high risk of bias or at high risk of bias, and we rated trials as low risk of bias overall
if all domains were rated as probably at low risk of bias or at low risk of bias. Reviewers resolved
discrepancies by means of discussion and, when necessary, a third reviewer served as arbiter.

Data synthesis
For dichotomous outcomes, we summarized the effect of interventions using odds ratios. When the
incidence of the outcome was low across studies (that is, there were no events in several study
groups), we used the risk difference. For continuous outcomes, we used the mean difference. When
studies reported the same outcome using a scale with a different range, we converted the data to the
most reported scale before conducting analyses. In addition, we calculated 95% CIs around all of
these estimates and created forest plots (in which the black diamond represents the pooled estimate
across studies). In instances when the SD was not reported, we calculated SDs using SE, CIs, means,
and sample sizes. In rare cases when none of these statistics were reported, we imputed the SD by
means of averaging the SDs of 3 studies with similar means. We conducted random-effects meta-
analyses using Review Manager (Cochrane) software.

Certainty of the evidence

We assessed the certainty of the evidence using the GRADE approach.9 Two methodologists
(F.V.-P., A.C.-L.) formally trained in using GRADE rated each domain for each comparison and

JADA 154(8) n http://jada.ada.org n August 2023 729

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 n = 240
Records identified in the
Living Overview of Evidence
platform for third-molar–related
literature n = 82
Records excluded on the basis of
abstract information for
population or intervention or
n = 240 study design
Records screened

n = 63
n = 103 Articles excluded on the basis of
Full-text articles assessed for eligibility full-text information for
(reported in 108 references) population or intervention
or comparison or outcome
Reasons for exclusion
• Wrong outcomes (n = 52)
• Wrong population (n = 3)
• Wrong intervention (n = 8)
n = 40
Included randomized controlled trial
(reported in 64 references)

Figure 1. Study identification and selection flowchart.

outcome independently, resolving discrepancies by means of discussion. We rated the certainty as

high, moderate, low, or very low, taking into consideration risk of bias, inconsistency, indirectness,
publication bias, and imprecision. We used a minimally contextualized approach with a null effect
threshold to rate the certainty that there is a benefit or harm.15 When the point estimate was close
to the null effect, we rated our certainty as having a trivial effect (that is, no important difference)
using a threshold of 10% of the baseline risk of dichotomous outcomes and 10% of the scale range
for continuous outcomes.16 For dichotomous outcomes, we calculated absolute estimates of effect
using the mean baseline risk across trials. We created GRADE summary of findings tables using
GRADEpro software (McMaster University and Evidence Prime).

Subgroup analyses
We performed 2 subgroup analyses to determine the extent to which treatment effects vary ac-
cording to the type of corticosteroid (for example, dexamethasone, methylprednisolone, prednis-
olone, and triamcinolone acetonide) and routes of administration (oral, intramuscular, submucosal)
compared with a placebo.

RESULTS
The search of the Epistemonikos database initially identified a total of 44 search results. Titles and
abstracts of all SRs were screened for our inclusion criteria, and 19 SRs proved relevant. These SRs
included a total of 79 RCTs (reported in 81 references) comparing the use of corticosteroids with a
placebo for patients undergoing surgical third-molar extractions. All identified RCTs were entered
into our database. The following link provides access to the interactive version of the matrix of
evidence that we built, as described above (corticosteroids vs a placebo for patients undergoing
third-molar extractions: http://www.epistemonikos.org/matrixes/60cfb16e7aaac86eee79456c).
After removal of duplicates, the search in the L$OVE platform for “corticosteroids” and “third
molar” yielded 240 records to screen at the title and abstract stage. Then, 103 of those records were
potentially eligible, and their full texts were evaluated. Finally, 40 RCTs (64 references) were
included.17-56 The complete study selection process, including the reasons for excluding studies at
the time of the full-text review, is summarized in a Preferred Reporting Items for Systematic Reviews
and Meta-Analyses flowchart (Figure 1).

730 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Table 1. Characteristics of the included studies examining corticosteroids.

PARTICIPANTS SEX,
STUDY RANDOMIZED, FEMALE, TYPE OF
STUDY* DESIGN COUNTRY NO. AGE, Y % EXTRACTION INTERVENTIONS
ElHag and Parallel England 70 Mean (SD), 51.42 Surgical tooth extraction 10 mg of dexamethasone 1 h
Colleagues,28 group 23.5 (NR†) (impacted third molar) preoperatively and 10-18 h
1985 postoperatively (intramuscular), no
corticosteroid treatment (placebo)

Pedersen and Split Denmark 66 Mean (SD), 22 56.66 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,47 mouth (NR) (impacted third molar) (intramuscular), no treatment
1985 (placebo)

Schmelzeisen Split Germany 80 Mean (SD), 18 45 Surgical tooth extraction 6 mg of dexamethasone 12 h

and mouth (NR) (impacted third molar) preoperatively and 12 h
Colleagues,51 postoperatively (oral), no treatment
1993 (placebo)

Buyukkurt and Parallel Turkey 30 Range, 18-36 44.44 Surgical tooth extraction 25 mg of prednisolone
Colleagues,23 group (impacted third molar) postoperatively (intramuscular), no
2006 treatment (placebo)

Grossi and Parallel Italy 61 Mean (SD), 45.9 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,33 group 27.7 (6.5) (impacted third molar) (submucosal), no treatment
2007A (placebo)

Grossi and Parallel Italy 61 Mean (SD), 45.9 Surgical tooth extraction 8 mg of dexamethasone
Colleagues,33 group 27.7 (6.5) (impacted third molar) (submucosal), no treatment
2007B (placebo)

Vegas- Parallel Spain 80 Mean (SD), 25 46 Surgical tooth extraction 40 mg of methylprednisolone

Bustamante and group (5) (impacted third molar) (intramuscular), no treatment
Colleagues,55
2008

Kang and Parallel Republic of Korea 450 Range, 20-30 NR Surgical tooth extraction 10 mg of prednisolone (oral), no
Colleagues,36 group (impacted third molar) treatment (placebo)
2010A

Kang and Parallel Republic of Korea 450 Range, 20-30 NR Surgical tooth extraction 20 mg of prednisolone (oral), no
Colleagues,36 group (impacted third molar) treatment (placebo)
2010B

Tiigimae and Parallel Estonia 78 Mean (SD) 73 Surgical tooth extraction 120 mg of etorikoxib
Colleagues,54 group 30.57 (13.73) (impacted third molar) preoperatively and 30 mg of
2010 prednisolone immediately
postoperatively (oral), 120 mg of
etorikoxib preoperatively

Antunes and Parallel Brazil 60 Mean (SD), 21 38.4 Surgical tooth extraction 8 mg of dexamethasone
Colleagues,18 group (5.4) (impacted third molar) (intramuscular), no treatment
2011A (placebo)

Antunes and Parallel Brazil 60 Mean (SD), 21 38.4 Surgical tooth extraction 8 mg of dexamethasone (oral), no
Colleagues18 group (5.4) (impacted third molar) treatment (placebo)
2011B

Deo and Parallel India 30 Mean (SD), 40 Surgical tooth extraction 8 mg of dexamethasone
Colleagues,25 group 24.93 (NR) (impacted third molar) (submucosal), no treatment
2011 (placebo)

Majid and Parallel Iraq 33 Mean (SD), 51.51 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,41 group 26.9 (6.1) (impacted third molar) (intramuscular), no treatment
2011A (placebo)

Majid and Parallel Iraq 33 Mean (SD), 51.51 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,41 group 26.9 (6.1) (impacted third molar) (submucosal), no treatment
2011B (placebo)

Majid and Parallel Iraq 30 Mean (SD), 46.66 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,42 group 26.7 (6.3) (impacted third molar) (intramuscular), no treatment
2011A (placebo)

Majid and Parallel Iraq 30 Mean (SD), 46.66 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,42 group 26.7 (6.3) (impacted third molar) (submucosal), no treatment
2011B (placebo)

* A, B, and C represent different interventions within the same study. † NR: Not reported.

JADA 154(8) n http://jada.ada.org n August 2023 731

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Table 1. Continued

PARTICIPANTS SEX,
STUDY RANDOMIZED, FEMALE, TYPE OF
STUDY* DESIGN COUNTRY NO. AGE, Y % EXTRACTION INTERVENTIONS
Bauer and Parallel Brazil 54 Mean (SD), 22 68.09 Surgical tooth extraction 600 mg of ibuprofen, 8 mg of
Colleagues,20 group (3.6) (impacted third molar) dexamethasone (oral), no
2012 corticosteroid treatment (placebo)

Bortoluzzi and Parallel Brazil 50 Mean (SD), NR Surgical tooth extraction 8 mg of dexamethasone and 2 g of
Colleagues,22 group 22.5 (NR) (impacted third molar) amoxicillin, 2 g of amoxicillin (oral)
2013A

Bortoluzzi and Parallel Brazil 50 Mean (SD), NR Surgical tooth extraction 8 mg of dexamethasone (oral), no
Colleagues,22 group 22.5 (NR) (impacted third molar) treatment (placebo)
2013B

Majid and Parallel Iraq 47 Mean (SD), 57.45 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,43 group 25.69 (5.53) (impacted third molar) (intramuscular), no treatment
2013A (placebo)

Majid and Parallel Iraq 47 Mean (SD), 57.45 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,43 group 25.69 (5.53) (impacted third molar) (submucosal), no treatment
2013B (placebo)

Majid and Parallel Iraq 47 Mean (SD), 57.45 Surgical tooth extraction 1 mg of dexamethasone (oral), no
Colleagues,43 group 25.69 (5.53) (impacted third molar) treatment (placebo)
2013C

Nair and Parallel India 100 NR NR Surgical tooth extraction 4 mg of dexamethasone

Colleagues,45 group (impacted third molar) (submucosal), no treatment
2013 (placebo)

Simone and Parallel Brazil 34 NR 70.59 Surgical tooth extraction 8 mg of dexamethasone (oral), no
Colleagues,53 group (impacted third molar) treatment (placebo)
2013

Bhargava and Parallel India 40 NR NR Surgical tooth extraction 4 mg/mL of dexamethasone

Colleagues,21 group (impacted third molar) (submucosal), no treatment
2014A (placebo)

Bhargava and Parallel India 40 NA NA Surgical tooth extraction 4 mg/mL of dexamethasone

Colleagues,21 group (impacted third molar) (intramuscular), no treatment
2014B (placebo)

Bhargava and Parallel India 40 NA NA Surgical tooth extraction 4 mg of dexamethasone (oral), no

Colleagues,21 group (impacted third molar) treatment (placebo)
2014C

Zerener and Parallel Turkey 78 Mean (SD), 48.72 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,56 group 22.6 (6.3) (impacted third molar) (submucosal), no treatment
2015A (placebo)

Zerener and Parallel Turkey 78 Mean (SD), 48.72 Surgical tooth extraction 4 mg of triamcinolone acetonide
Colleagues,56 group 22.6 (6.3) (impacted third molar) (submucosal), no treatment
2015B (placebo)

Deo and Parallel Nepal 40 Mean (SD), 36.67 Surgical tooth extraction 2 mL of 4 mg/mL of dexamethasone
Colleagues,26 group 24.93 (NR) (impacted third molar) (submucosal), no treatment
2016A (placebo)

Deo and Parallel Nepal 40 Range, 20-41 NR Surgical tooth extraction 2 mL of 4 mg/mL of dexamethasone
Colleagues,27 group (impacted third molar) (submucosal), no treatment
2016B (placebo)

Ibikunle and Parallel Nigeria 191 Mean (SD), 62.9 Surgical tooth extraction 40 mg of prednisolone (oral), no
Colleagues,34 group 28.1 (7.4) (impacted third molar) treatment (placebo)
2016A

Ibikunle and Parallel Nigeria 191 Mean (SD), 62.9 Surgical tooth extraction 40 mg of prednisolone
Colleagues,34 group 28.1 (7.4) (impacted third molar) (submucosal), no treatment
2016B (placebo)

Prashar and Parallel India 30 Mean (SD), NR Surgical tooth extraction 8 mg of methylprednisolone and 50
Colleagues,48 group 26.5 (NR) (impacted third molar) mg of diclofenac (oral), 50 mg of
2016 diclofenac

Saravanan and Parallel India 60 NR NR Surgical tooth extraction 4 mg of dexamethasone

Colleagues,50 groups (impacted third molar) (intramuscular), no treatment
2016A (placebo)

732 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Table 1. Continued

PARTICIPANTS SEX,
STUDY RANDOMIZED, FEMALE, TYPE OF
STUDY* DESIGN COUNTRY NO. AGE, Y % EXTRACTION INTERVENTIONS
Saravanan and Parallel India 60 NR NR Surgical tooth extraction 4 mg of dexamethasone
Colleagues,50 groups (impacted third molar) (submucosal), no treatment
2016B (placebo)

Ghensi and Parallel Italy 80 Mean (SD), 27 46.25 Surgical tooth extraction 4 mg of dexamethasone
Colleagues,29 group (7.1) (impacted third molar) (submucosal), no treatment
2017A (placebo)
Ghensi and Parallel Italy 80 Mean (SD), 27 46.25 Surgical tooth extraction 4 mg of dexamethasone and 40 mg
Colleagues,29 group (7.1) (impacted third molar) of bromelain (submucosal), 40 mg of
2017B bromelain

Gopinath and Parallel India 80 NR NR Surgical tooth extraction 4 mg of dexamethasone

Colleagues,31 group (impacted third molar) (submucosal), no treatment
2017 (placebo)
Gozali and Split Thailand 96 Range, 18-30 60.42 Surgical tooth extraction 8 mg of dexamethasone
Colleagues,32 mouth (impacted third molar) (submucosal), no treatment
2017 (placebo)

Khalida and Parallel Pakistan 60 Mean (SD), 36.67 Surgical tooth extraction 4 mg/mL of dexamethasone
Colleagues,37 group 28.77 (7.04) (impacted third molar) (submucosal), no treatment
2017 (placebo)
Lim and Parallel Malaysia 65 Mean (SD), 25 81.67 Surgical tooth extraction 4 mg/mL of dexamethasone
Colleagues,40 group (4) (impacted third molar) (submucosal), no treatment
2017A (placebo)

Lim and Parallel Malaysia 65 Mean (SD), 25 81.67 Surgical tooth extraction 40 mg of methylprednisolone
Colleagues,40 group (4) (impacted third molar) (submucosal), no treatment
2017B (placebo)
Mojsa and Parallel Poland 60 Range, 18-42 64.44 Surgical tooth extraction 4 mg/mL of dexamethasone
Colleagues,44 group (impacted third molar) preoperatively (submucosal), no
2017A treatment (placebo)

Mojsa and Parallel Poland 60 Range, 18-42 64.44 Surgical tooth extraction 4 mg/mL of dexamethasone
Colleagues,44 group (impacted third molar) postoperatively (submucosal), no
2017B treatment (placebo)
Selimovic and Parallel Bosnia and 40 Range, 18-45 NR Surgical tooth extraction 32 mg of methylprednisolone and
Colleagues,52 group Herzegovina (impacted third molar) 15 mg of meloxicam (oral), 15 mg of
2017 meloxicam

Afkan and Parallel Iran 75 Mean (SD), 28 NR Surgical tooth extraction 2 mg of dexamethasone
Colleagues,17 group (NR) (impacted third molar) preoperatively and postoperatively
2018A (oral), no treatment (placebo)
Afkan and Parallel Iran 75 Mean (SD), 28 NR Surgical tooth extraction 2 mg of dexamethasone (oral), no
Colleagues,17 group (NR) (impacted third molar) treatment (placebo)
2018B

Chugh and Parallel India 60 Mean (SD), 36.67 Surgical tooth extraction 8 mg of dexamethasone
colleagues,24 group 29.79 (8.37) (impacted third molar) (submucosal), no treatment
2018A (placebo)
Chugh and Parallel India 60 Mean (SD), 36.67 Surgical tooth extraction 40 mg of methylprednisolone
colleagues,24 group 29.79 (8.37) (impacted third molar) (submucosal), no treatment
2018B (placebo)

Atalay and Parallel Turkey 60 Mean (SD), 46.67 Surgical tooth extraction 1 mL of 4 mg/mL of dexamethasone
colleagues,19 group 25.18 (5.26) (impacted third molar) solution (submucosal), no treatment
2020A (placebo)
Atalay and Parallel Turkey 60 Mean (SD), 46.67 Surgical tooth extraction 2 mL of 8 mg/mL of examethasone
colleagues,19 group 25.18 (5.26) (impacted third molar) solution (submucosal), no treatment
2020B (placebo)

Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 20 mg of methylprednisolone
Colleagues,38 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2020A (placebo)
Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 30 mg of methylprednisolone
Colleagues,38 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2020B (placebo)

JADA 154(8) n http://jada.ada.org n August 2023 733

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Table 1. Continued

PARTICIPANTS SEX,
STUDY RANDOMIZED, FEMALE, TYPE OF
STUDY* DESIGN COUNTRY NO. AGE, Y % EXTRACTION INTERVENTIONS
Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 40 mg of methylprednisolone
Colleagues,38 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2020C (placebo)

Pansard and Parallel Brazil 114 Mean (SD), 65.59 Surgical tooth extraction 8 mg of dexamethasone (oral), no
Colleagues,46 group 31.43 (11.76) (impacted third molar) treatment (placebo)
2020

Sahu and Parallel India 40 Range, 18-60 NR Surgical tooth extraction 4 mg/mL of dexamethasone
Colleagues,49 group (impacted third molar) (submucosal), no treatment
2020 (placebo)

Gholami and Parallel Iran 60 Mean (SD), 51.67 Surgical tooth extraction 40 mg of methylprednisolone
Colleagues,30 group 26.83 (4.19) (impacted third molar) (intramuscular, masseter), no
2021A treatment (placebo)

Gholami and Parallel Iran 60 Mean (SD), 51.67 Surgical tooth extraction 40 mg of methylprednisolone
Colleagues,30 group 26.83 (4.19) (impacted third molar) (intramuscular, gluteal), no
2021B treatment (placebo)

Imon and Parallel Bangladesh 294 Mean (SD), 25 44.9 Surgical tooth extraction Dexamethasone tapering dose (oral),
Colleagues,35 group (NR) (impacted third molar) no treatment (placebo)
2021

Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 20 mg of methylprednisolone
Colleagues,39 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2021A (placebo)

Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 30 mg of methylprednisolone
Colleagues,39 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2021B (placebo)

Larsen and Split Denmark 104 Mean (SD), 69.23 Surgical tooth extraction 40 mg of methylprednisolone
Colleagues,39 mouth 25.92 (5.99) (impacted third molar) (intramuscular), no treatment
2021C (placebo)

Characteristics of included studies

Most RCTs were parallel group in design (88%), with the number of participants ranging from 30
through 450. Mean (SD) age of participants across studies ranged from 18 (not reported) to 31.42
(11.76) years. All populations across included studies underwent surgical extraction of the impacted
third molars (Table 1).

Risk of bias in included studies

Randomization, deviations from the intended intervention, and measurement of outcomes were the
risk of bias domains judged as high or probably high risk of bias most frequently across the included
studies (eTable 2, available online at the end of this article).

Subgroup analyses
The various types of corticosteroids identified in this review included dexamethasone, methyl-
prednisolone, prednisolone, and triamcinolone acetonide. The route of administration varied
among oral, intramuscular, and submucosal. We did not find evidence of a subgroup effect for the
comparison of each specific corticosteroid with a placebo (Figures 2-4). Similarly, we did not find
compelling evidence of a subgroup effect for the comparison of corticosteroids with a placebo based
on administration modes (Figures 3 and 4). Any minor quantitative differences observed across
analyses were not clinically significant.

Outcome measures
Corticosteroids Compared With a Placebo (No Treatment With Corticosteroids)
Postoperative pain
Seven RCTs, including 396 participants, assessed pain using a visual analog scale ranging from
0 (no pain) through 100 (worst pain imaginable) at 6 hours.17,20,22,23,44,53,55 The results suggested

734 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Mean difference Mean difference
Corticosteroid Placebo Weight, inverse variance, inverse variance,
Study or subgroup Mean SD Total Mean SD Total % random, 95% CI random, 95% CI

1.1.1 Dexamethasone
Bauer and Colleagues,20 2012 16 19 24 22 19 23 16.1 –6.00 (–16.87 to 4.87)
Bortoluzzi and Colleagues,22 2013A 28.7 19 12 25.25 26.055 12 8.2 3.45 (–14.80 to 21.70)
Bortoluzzi and Colleagues,22 2013B 32.7 19 14 28.2 26.055 12 8.6 4.50 (–13.29 to 22.29)
53
Simone and Colleagues, 2013 18.3 19 20 36.7 26.055 14 10.0 –18.40 (–34.39 to –2.41)
44
Mojsa and Colleagues, 2017A 23.89 9.59 30 33.42 43.88 15 5.9 –9.53 (–32.00 to 12.94)
44
Mojsa and Colleagues, 2017B 7.08 17.14 30 33.42 43.88 15 5.7 –26.34 (–49.38 to –3.30)
Afkan and Colleagues,17 2018A 77.2 22.27 25 80.8 21.78 12 10.8 –3.60 (–18.70 to 11.50)
17 2018B
Afkan and Colleagues, 80.4 21.89 25 80.8 21.78 13 11.3 –0.40 (–15.02 to 14.22)
Subtotal (95% CI) 180 116 76.6 –5.96 (–12.27 to 0.35)
Heterogeneity: τ2 = 13.99, χ 27 = 8.43 (P = .30), I 2 = 17%
Test for overall effect: z = 1.85 (P = .06)
1.1.2 Methylprednisolone
Vegas-Bustamante and Colleagues,55 2008 13.66 18.38 35 31.44 26.055 35 16.6 –17.78 (–28.34 to –7.22)
Subtotal (95% CI) 35 35 16.6 –17.78 (–28.34 to –7.22)
Heterogeneity: Not applicable
Test for overall effect: z = 3.30 (P = .0010)

1.1.3 Prednisolone
Buyukkurt and Colleagues,23 2006 31.93 26.87 15 49.6 30.33 15 6.9 –17.67 (–38.18 to 2.84)
Subtotal (95% CI) 15 15 6.9 –17.67 (–38.18 to 2.84)
Heterogeneity: Not applicable
Test for overall effect: z = 1.69 (P = .09)

Total (95% CI) 230 166 100.0 –8.79 (–14.80 to –2.77)

Heterogeneity: τ2 = 27.73, χ 29 = 12.96 (P = .16), I 2 = 31%
Test for overall effect: z = 2.86 (P = .004) –50 –25 0 25 50
Test for subgroup differences: χ 22 = 4.18, (P = .12), I 2 = 52.1% Favors corticosteroid Favors placebo

Figure 2. Corticosteroids vs a placebo for the outcome of pain at 6 hours (corticosteroid type). A and B represent different interventions within the same
study.

that there may be a trivial benefit of corticosteroids compared with a placebo in terms of pain
intensity measured 6 hours postoperatively (mean difference, 8.79 points lower; 95% CI, 14.8 to
2.77 points lower; low certainty) (Table 2, Figure 2). Twenty-three RCTs, including 1,555 par-
ticipants, assessed pain using a visual analog scale ranging from 0 (no pain) through 100 (worst pain
imaginable) at 24 hours.18,20-22,24-26,30,32,34,36,39-44,48-50,53,55,56 Very low certainty evidence sug-
gested that there may be a trivial difference between corticosteroids and a placebo in terms of pain
intensity measured 24 hours postoperatively (mean difference, 8.89 points lower; 95% CI 10.71 to
7.06 points lower; very low certainty) (Table 2, Figure 4).

Reported Adverse Effects

Infection
Twenty-one RCTs, including 1,630 participants, assessed the incidence of postoperative infection
as an adverse effect at different follow-up times.19,22,24,28,29,31,33,35,37,38,40-47,51,54,55 The studies
suggested that there may be no difference between corticosteroids and a placebo with regard to
incidence of postoperative infection (risk difference, 0%; 95% CI, –1% to 1%; low certainty)
(Table 2, eFigure 1, available online at the end of this article).

Alveolar osteitis
Eight RCTs, including 410 participants, examined the occurrence of alveolar osteitis as the pro-
portion of participants experiencing this outcome at any time.22,24,28,37,41-43,47 Very low certainty
evidence suggested that there may be no difference between corticosteroids and a placebo in terms
of incidence of alveolar osteitis (risk difference, 0%; 95% CI, –3% to 4%; very low certainty)
(Table 2, eFigure 2, available online at the end of this article).

GI adverse effects
Three RCTs, including 120 participants, provided evidence on the incidence of GI adverse effects
expressed as the proportion of participants experiencing nausea or vomiting at any time.27,51,52
Corticosteroids may reduce the incidence of GI adverse effects compared with a placebo; howev-
er, the evidence is uncertain (risk difference, –19%; 95% CI, –54% to 16%; very low certainty)
(Table 2, eFigure 3, available online at the end of this article). In addition, although we identified a
statistically significant difference in GI adverse effects depending on the route of administration

JADA 154(8) n http://jada.ada.org n August 2023 735

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Mean difference Mean difference
Corticosteroid Placebo Weight, inverse variance, inverse variance,
Study or subgroup Mean SD Total Mean SD Total % random, 95% CI random, 95% CI

1.8.1 Intramuscular
Buyukkurt and Colleagues,23 2006 31.93 26.87 15 49.6 30.33 15 6.9 –17.67 (–38.18 to 2.84)
Vegas-Bustamante and Colleagues,55 2008 13.66 18.38 35 31.44 26.055 35 16.6 –17.78 (–28.34 to –7.22)
Subtotal (95% CI) 50 50 23.4 –17.76 (–27.15 to –8.37)
Heterogeneity: 2 = 0.00,  21 = 0.00 (P = .99), I 2 = 0%
Test for overall effect: z = 3.71 (P = .0002)

1.8.2 Oral
Bauer and Colleagues,20 2012 16 19 24 22 19 23 16.1 –6.00 (–16.87 to 4.87)
Bortoluzzi and Colleagues,22 2013A 28.7 19 12 25.25 26.055 12 8.2 3.45 (–14.80 to 21.70)
22
Bortoluzzi and Colleagues, 2013B 32.7 19 14 28.2 26.055 12 8.6 4.50 (–13.29 to 22.29)
Simone and Colleagues, 201353 18.3 19 20 36.7 26.055 14 10.0 –18.40 (–34.39 to –2.41)
17
Afkan and Colleagues, 2018A 77.2 22.27 25 80.8 21.78 12 10.8 –3.60 (–18.70 to 11.50)
Afkan and Colleagues,17 2018B 80.4 21.89 25 80.8 21.78 13 11.3 –0.40 (–15.02 to 14.22)
Subtotal (95% CI) 120 86 65.0 –4.20 (–10.22 to 1.82)
Heterogeneity: 2 = 0.00,  25 = 5.00 (P = .42), I 2 = 0%
Test for overall effect: z = 1.37 (P = .17)

1.8.3 Submucosal
Mojsa and Colleagues,44 2017A 23.89 9.59 30 33.42 43.88 15 5.9 –9.53 (–32.00 to 12.94)
Mojsa and Colleagues,44 2017B 7.08 17.14 30 33.42 43.88 15 5.7 –26.34 (–49.38 to –3.30)
Subtotal (95% CI) 60 30 11.6 –17.73 (–34.20 to –1.27)
Heterogeneity: 2 = 6.50,  21 = 1.05 (P = .31), I 2 = 5%
Test for overall effect: z = 2.11 (P = .03)

Total (95% CI) 230 166 100.0 –8.79 (–14.80 to –2.77)

Heterogeneity: 2 = 27.73,  29 = 12.96 (P = .16), I 2 = 31%
Test for overall effect: z = 2.86 (P = .004) –50 –25 0 25 50
Test for subgroup differences:  22 = 6.87, (P = .03), I 2 = 70.9% Favors corticosteroid Favors placebo

Figure 3. Corticosteroids vs a placebo for the outcome of pain at 6 hours (route of administration). A and B represent different interventions within the
same study.

(oral multidose vs submucosal single dose), the small number of events and studies did not provide
compelling evidence to claim a subgroup effect (Table 2, eFigure 4, eFigure 5, available online at
the end of this article).

Total pain relief, global efficacy rating, and mood alteration

None of the included studies provided evidence related to the effect of corticosteroids compared
with a placebo on total pain relief, global efficacy rating, or mood alteration outcomes.

DISCUSSION
Low certainty evidence suggested that corticosteroids administered orally, submucosally, or intra-
muscularly in adolescent, adult, or older adult participants may decrease pain intensity a trivial
amount at 6 hours compared with a placebo (no treatment with corticosteroids). This difference
remained trivial at 24 hours (very low certainty). We found no difference between corticosteroids
compared with a placebo with regard to the incidence of postoperative infection (low certainty) and
alveolar osteitis (very low certainty). With regard to adverse events, in particular GI events (for
example, nausea and vomiting), the available evidence suggested that corticosteroids could result in
a small benefit (that is, a reduction in GI adverse events) compared with a placebo (very low
certainty); however, these results are not trustworthy due to the small number of studies and par-
ticipants. Most of the included RCTs had serious issues related to the randomization process, de-
viations from the intended intervention, and measurement of outcomes. In addition, none of the
primary studies were conducted in the United States, where intravenous corticosteroids (not oral,
submucosal, intramuscular) are often administered, patients regularly receive intravenous sedation,
and more than 1 mandibular third molar may be extracted at a single appointment. To improve the
certainty of the evidence, future trials should focus on conducting methodologically rigorous RCTs
and increasing the sample size.
In an SR including 12 RCTs, researchers examined the impact of corticosteroids on post-
operative morbidity after third-molar extraction.4 The researchers examined the effects of
perioperative corticosteroid administration (that is, betamethasone, prednisolone, methylpred-
nisolone, methylprednisolone acetate, methylprednisolone sodium succinate, dexamethasone)
on pain intensity, trismus, and swelling. Their finding suggested that corticosteroids reduced

736 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Mean difference, Mean difference,
Corticosteroid Placebo Weight, inverse variance, inverse variance,
Study or subgroup Mean SD Total Mean SD Total % random 95% CI random 95% CI

1.7.1 Submucosal dexamethasone

Antunes and Colleagues,18 2011A 1.4 1.1 18 2.1 1.02 11 4.8 –0.70 (–1.49 to 0.09)
Deo and Colleagues,25 2011 61.5 12.61 19 64.5 12.795 11 2.1 –3.00 (–12.45 to 6.45)
Majid and Colleagues,41 2011B 2.8 2.9 11 6.1 2.8 5 4.3 –3.30 (–6.29 to –0.31)
Majid and Colleagues,42 2011B 2.8 3.1 10 6.3 2.9 5 4.2 –3.50 (–6.69 to –0.31)
Majid and Colleagues,43 2013B 29 27 11 70 20 4 0.5 –41.00 (–66.27 to –15.73)
Bhargava and Colleagues,21 2014A 18 7 10 30 6 4 2.7 –12.00 (–19.31 to –4.69)
Zerener and Colleagues,56 2015A 11.5 13.51 26 24.19 17.75 13 1.8 –12.69 (–23.65 to –1.73)
Deo and Colleagues,26 2016 68.44 11.22 19 71.67 16.67 11 1.7 –3.23 (–14.30 to 7.84)
Saravanan and Colleagues,50 2016B 26.4 4.84 20 66.3 8.47 10 3.3 –19.90 (–45.56 to –34.24)
Gozali and Colleagues,32 2017 14.12 7.81 48 23.85 12.73 48 3.8 –9.73 (–13.96 to –5.50)
Lim and Colleagues,40 2017A 28.1 6.94 20 36.2 12.795 10 2.4 –8.10 (–16.59 to 0.39)
Mojsa and Colleagues,44 2017A 11.34 6.94 30 13.06 12.86 15 2.8 –1.72 (–8.69 to 5.25)
Mojsa and Colleagues,44 2017B 0 1.84 30 13.06 12.86 15 3.0 –13.06 (–19.60 to –6.52)
Chugh and Colleagues,24 2018A 35 31.3 23 58 37.8 9 0.4 –23.00 (–50.81 to 4.81)
Sahu and Colleagues,49 2020 12 7.6 20 27.5 11.6 20 3.1 –15.50 (–21.58 to –9.42)
Subtotal (95% CI) 315 191 40.9 –10.91 (–16.02 to –5.81)
2
Heterogeneity: τ2 = 80.97, χ = 249.66 (P < .00001), I 2 = 94%
14
Test for overall effect: z = 4.19 (P < .0001)

1.7.2 Oral dexamethasone

Antunes and Colleagues,18 2011B 1.65 1.35 20 2.1 1.02 11 4.8 –0.45 (–1.29 to 0.39)
Bauer and Colleagues,20 2012 15 14 24 19 18 23 2.2 –4.00 (–13.25 to 5.25)
Bortoluzzi and Colleagues,22 2013A 25.53 12.61 12 24.62 12.795 12 1.9 0.91 (–9.25 to 11.07)
Bortoluzzi and Colleagues,22 2013B 23.62 12.61 14 15.38 12.795 12 2.0 8.24 (–1.56 to 18.4)
Majid and Colleagues,43 2013C 30 26 12 70 20 4 0.5 –40.00 (–64.51 to –15.49)
Simone and Colleagues,53 2013 15 12.61 20 22.5 12.795 14 2.3 –7.50 (–16.19 to 1.19)
Bhargava and Colleagues,21 2014C 19 12 10 30 6 3 2.0 –11.00 (–21.07 to –0.93)
Subtotal (95% CI) 112 79 15.6 –3.81 (–9.43 to 1.81)
2
Heterogeneity: τ2 = 33.68, χ = 20.30 (P = .002), I 2 = 70%
6
Test for overall effect: z = 1.33 (P = .18)
1.7.3 Intramuscular dexamethasone
Majid and Colleagues,41 2011A 3.7 2.9 11 6.1 2.8 6 4.3 –2.40 (–5.22 to 0.42)
Majid and Colleagues,42 2011A 3.6 3.1 10 6.3 2.9 5 4.2 –2.70 (–5.89 to 0.49)
Majid and Colleagues,43 2013A 32 30 12 70 20 4 0.5 –38.00 (–63.93 to –12.07)
Bhargava and Colleagues,21 2014B 16 8 10 30 6 3 2.4 –14.00 (–22.1 to –5.59)
50
Saravanan and Colleagues, 2016A 29.75 9.66 20 66.3 8.47 10 2.9 –36.55 (–43.29 to –29.81)
Subtotal (95% CI) 63 28 14.3 –16.24 (–27.96 to –4.52)
2
Heterogeneity: τ2 = 152.04, χ = 98.49 (P < .00001), I 2 = 96%
4
Test for overall effect: z = 2.72 (P = .007)

1.7.4 Submucosal methylprednisolone

Lim and Colleagues,40 2017B 21.9 24.25 20 36.2 12.795 10 1.4 –14.30 (–27.56 to –1.04)
Chugh and Colleagues,24 2018B 53 32.5 20 58 37.8 8 0.3 –5.00 (–34.82 to 24.82)
Subtotal (95% CI) 40 18 1.7 –12.76 (–24.88 to –0.65)
2
Heterogeneity: τ2 = 0.00, χ = 0.31 (P = .58), I 2 = 0%
1
Test for overall effect: z = 2.06 (P = .04)
1.7.5 Oral methylprednisolone
Prashar and Colleagues,48 2016 26.26 14.64 15 50 18.92 15 1.5 –23.74 (–35.85 to –11.63)
Subtotal (95% CI) 15 15 1.5 –23.74 (–35.85 to –11.63)
Heterogeneity: Not applicable
Test for overall effect: z = 3.84 (P = .0001)

1.7.6 Intramuscular methylpredmisolone

Vegas-Bustamante and Colleagues,55 2008 8.97 24.25 35 19.59 12.795 35 2.2 –10.62 (–19.70 to –1.54)
Gholami and Colleagues,30 2021A 67.5 21.7 20 54 25.8 10 0.8 13.50 (–5.11 to 32.11)
Gholami and Colleagues,30 2021B 47 26.8 20 54 25.8 10 0.7 –7.00 (–26.84 to 12.84)
Larsen and Colleagues,39 2021A 40.25 8.58 26 57.17 10.71 9 2.6 –16.92 (–24.66 to –9.18)
Larsen and Colleagues,39 2021B 37.88 6.21 26 57.17 10.71 9 2.7 –19.29 (–26.68 to –11.90)
Larsen and Colleagues,39 2021C 45.17 7.67 26 57.17 10.71 8 2.5 –12.00 (–19.99 to –4.01)
Subtotal (95% CI) 153 81 11.5 –11.80 (–18.23 to –5.36)
2
Heterogeneity: τ = 34.82, χ = 12.13 (P = .03), I = 59%
2 2
5
Test for overall effect: z = 3.59 (P = .0003)

1.7.7 Submucosal prednisolone

Ibikunle and Colleagues,34 2016B 6.44 1.93 62 8.2 1.54 31 4.8 –1.76 (–2.48 to –1.04)
Subtotal (95% CI) 62 31 4.8 –1.76 (–2.48 to –1.04)
Heterogeneity: Not applicable
Test for overall effect: z = 4.76 (P < .00001)
1.7.8 Oral prednisolone
Kang and Colleagues,36 2010A 66.3 34.4 60 70.3 28.1 48 1.6 –4.00 (–15.79 to 7.79)
Kang and Colleagues,36 2010B 62.8 27.5 64 70.3 28.1 48 1.9 –7.50 (–17.92 to –2.92)
Ibikunle and Colleagues,34 2016A 6.34 1.95 62 8.2 1.54 31 4.8 –1.86 (–2.59 to –1.13)
Subtotal (95% CI) 186 127 8.3 –1.90 (–2.62 to –1.17)
2
Heterogeneity: τ2 = 0.00, χ = 1.24 (P = .54), I 2 = 0%
2
Test for overall effect: z = 5.13 (P < .00001)
1.7.9 Submucosal triamcimolone acetonide
Zerener and Colleagues,56 2015B 15.76 21.03 26 24.19 17.75 13 1.5 –8.43 (–21.02 to 4.16)
Subtotal (95% CI) 26 13 1.5 –8.43 (–21.02 to 4.16)
Heterogeneity: Not applicable
Test for overall effect: z = 1.31 (P = .19)
Total (95% CI) 972 583 100.0 –8.89 (–10.71 to –7.06)
2
Heterogeneity: τ2 = 18.09, χ = 465.10 (P < .00001), I 2 = 91%
40
Test for overall effect: z = 9.53 (P < .00001) –50 –25 0 25 50
2
Test for subgroup difference: χ = 43.48, (P < .00001), I 2 = 81.6% Favors corticosteroid Favors placebo
8

Figure 4. Corticosteroids vs a placebo for the outcome of pain at 24 hours (corticosteroid type and route of administration). A, B, and C represent
different interventions within the same study.

swelling and trismus in the early (1-3 days) and late (> 3 days) postoperative phases. The re-
searchers could not determine the effect of corticosteroids on pain intensity. This differed from
our findings, which suggested a negligible difference between corticosteroids and a placebo at 6
and 24 hours postoperatively. In another SR, researchers also examined the effect of cortico-
steroids and included 28 RCTs.5 The researchers concluded that corticosteroids (that is,

JADA 154(8) n http://jada.ada.org n August 2023 737

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Table 2. Corticosteroids vs a placebo (no treatment with corticosteroids) for acute dental pain.

OUTCOME RELATIVE ANTICIPATED ABSOLUTE EFFECTS CERTAINTY† WHAT HAPPENS

EFFECT
(95% CI)* With No
Treatment With Difference
(Placebo) Corticosteroids (95% CI)
Pain

Assessed with visual analog scale –‡ Median pain, – Mean difference 8.79 Low{,# There may be a trivial benefit of
ranging from 0 (no pain) through 32.43 points points lower (14.8 to corticosteroids compared with no
100 (worst pain) 2.77 points lower) treatment (placebo) in terms of
Follow-up: 6 h pain measured 6 h
No. of participants: 396 (7 RCTs§) postoperatively.

Assessed with visual analog scale ranging – Median pain, – Mean difference 8.89 Very There is very low certainty
from 0 (no pain) through 100 (worst 26.06 points points lower (10.71 to low**,††,‡‡ evidence regarding the difference
pain) 7.06 points lower) between corticosteroids and no
Follow-up: 24 h treatment (placebo) in terms of
No. of participants: 1,555 (23 RCTs) pain measured 24 h
postoperatively.

Adverse Effect

Postoperative infection Not estimable 1.3% 2.4% 0% fewer (1% fewer Low§§,{{ There may be no difference
Assessed with proportion of patients to 1% more) between corticosteroids and no
experiencing postoperative infection treatment (placebo) with regard to
Follow-up: any time incidence of postoperative
No. of participants: 1,630 (21 RCTs) infection.

Alveolar osteitis Not estimable 1.2% 0.8% 0% fewer (3% fewer Very There is very low certainty
Assessed with proportion of patients to 4% more) low§§,## evidence regarding the difference
experiencing alveolar osteitis between corticosteroids and no
Follow-up: any time treatment (placebo) in terms of
No. of participants: 410 (8 RCTs) incidence of alveolar osteitis.

Gastrointestinal Not estimable 21.4% 7.8% 19% fewer (54% fewer Very There is very low certainty
Assessed with proportion of patients to 16% more) low***,†††,‡‡‡ evidence regarding the difference
experiencing nausea/vomiting between corticosteroids and no
Follow-up: any time treatment (placebo) in terms of
No. of participants: 120 (3 RCTs) incidence of nausea/vomiting.

Total Pain Relief at 6 h, – – – – – –

Not Measured

Global Efficacy Rating at 6 h, Not – – – – – –

Measured

Mood Alteration, Not Measured – – – – – –

* The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
† GRADE Working Group grades of evidence: High certainty: very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty:
moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low certainty: confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: very little
confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect. ‡ –: No data were available for the outcome. § RCT:
Randomized controlled trial. { Four studies were at a high risk of reporting bias as they did not provide measures of variability. Three studies were at either a high or
probably high risk of attrition bias due to missing outcome data. Therefore, the authors rated down 1 level due to risk of bias. # Using a threshold of 10 points (based
on 10% of the scale range), the lower bound of the CI suggests an important difference favoring corticosteroids and the upper bound suggests a trivial difference
favoring corticosteroids. Therefore, the authors rated down 1 level due to imprecision. ** Several studies were probably at a high risk of bias arising from the
randomization process because there was no mention of allocation concealment and the health care providers were not blinded or it was unclear whether they were
blinded. Several studies were also at a high risk of performance and detection bias due to a lack of blinding of participants (outcome assessors). In addition, some
studies were at a high risk of reporting bias, as they did not report measures of variability. Therefore, the authors rated down 1 level due to risk of bias. †† There is
high statistical heterogeneity (I2 ¼ 91%; P < .00001) and the effect estimates of some studies are importantly different from each other. Therefore, the authors rated
down 1 level due to inconsistency. ‡‡ Using a threshold of 10 points (based on 10% of the scale range), the lower bound of the CI suggests an important difference
favoring corticosteroids and the upper bound suggests a trivial difference favoring corticosteroids. Therefore, the authors rated down 1 level due to imprecision.
§§ Several studies were probably at a high risk of bias arising from the randomization process because there was no mention of allocation concealment and the health
care providers were not blinded or it was unclear whether they were blinded. Several studies were also at a high risk of bias due to a lack of blinding of participants.
Therefore, the authors rated down 1 level due to risk of bias. {{ Given the very low event rate, the optimal information size for this outcome was not met. Therefore,
the authors rated down 1 level due to imprecision. ## Using a threshold of 0.12% (based on 10% of the baseline risk, that is, the risk with a placebo), the lower
bound of the CI suggests an important difference favoring corticosteroids and the upper bound suggests an important difference favoring no treatment (placebo).
Therefore, the authors rated down 2 levels due to imprecision. *** Two studies were at a high risk of attrition bias due to missing outcome data. Two studies were at
a high or probably high risk of detection bias as participants were not blinded. One of these studies also had concerns regarding reporting bias as well as the
randomization process, as it did not mention allocation concealment and it was unclear whether health care providers were blinded. Therefore, the authors rated
down 1 level due to risk of bias. ††† There is high statistical heterogeneity (I2 ¼ 90%; P < .0001) and the CI of 1 study that contributes 29.7% to the pooled estimate
does not overlap with the others. Therefore, the authors rated down 1 level due to inconsistency. ‡‡‡ Using a threshold of 2.14% (based on 10% of the baseline risk,
that is, the risk with a placebo), the lower bound of the CI suggests an important difference favoring corticosteroids and the upper bound suggests an important
difference favoring no treatment (placebo). Therefore, the authors rated down 1 level due to imprecision.

738 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
dexamethasone, prednisolone, methylprednisolone, betamethasone) improved patients’ post-
operative experiences and had a significant benefit on trismus and inflammation. As we focused
our review on pain intensity and adverse effects, we did not examine the evidence regarding
trismus, swelling, or inflammation; therefore, we cannot compare the results of the reviews cited
above with the results of our review.
In a third SR and meta-analysis, including 8 RCTs, researchers assessed the effect of submucosal
injection of dexamethasone after third-molar extraction.6 The results of the review suggested that
the submucosal injection of dexamethasone resulted in a reduction in swelling and pain from
impacted third-molar surgery. In that review, the researchers also found no difference between
dexamethasone and a placebo in relation to trismus. These results are similar to those from our
review in terms of reduction of pain with the use of corticosteroids.
In a fourth SR, including 7 RCTs, researchers examined the impact of different dosages of cortico-
steroids (that is, cortisol, prednisone, prednisolone, methylprednisolone, dexamethasone, and beta-
methasone) and administration routes on facial swelling, pain, and trismus.7 The researchers concluded
that preoperative submucosal injection of corticosteroids significantly diminished facial swelling,
postoperative pain, and trismus compared with a placebo. Although we did not examine facial swelling
and trismus, our findings are aligned regarding pain reduction, although the magnitude of effect that we
found was lower than our clinically significant threshold (10%). The researchers concluded that the
optimal dosage of corticosteroids and administration route for decreasing postsurgical morbidity and
improving quality of life after surgical removal of mandibular third molars was unknown.7
In a fifth SR, researchers analyzed the efficacy of corticosteroids for pain management after
mandibular third-molar extraction.8 The review included 27 RCTs and the researchers concluded
that corticosteroids (that is, dexamethasone, methylprednisolone, and betamethasone) could be used
as an adjuvant for pain reduction after an impacted third-molar extraction. The researchers also
suggested that methylprednisolone and dexamethasone delivered via an intramuscular route were the
best interventions for effective pain reduction. The ideal time for administration of corticosteroids was
the preoperative period. These results were similar to the results of our review in terms of reduction of
pain. However, our review included dexamethasone, methylprednisolone, prednisolone, and triam-
cinolone acetonide and did not find a difference in pain intensity among the different corticosteroids.
The researchers in the SRs mentioned did not assess the certainty of the evidence. The results of
our SR established that the best evidence supporting the outcomes of choice ranged from low
through very low certainty, highlighting the need for RCTs of higher methodological and reporting
quality, as well as statistical precision.

Strengths and limitations

The strengths of our review relied on each stage of the process being conducted independently and in
duplicate. We assessed the risk of bias for each RCT and the certainty of the evidence for each outcome
of interest. We performed the analyses and interpreted the results using the latest methodological
guidance from the GRADE working group. A limitation is that we restricted our eligibility criteria to
peer-reviewed research articles published in English. However, we believe it is unlikely that our
conclusions would have been different had we included studies in a different language.

CONCLUSIONS
The results of our SR and meta-analysis indicated that there was low and very low certainty evidence
informing the effect of corticosteroids administered orally, submucosally, or intramuscularly in adoles-
cent, adult, or older adult participants compared with a placebo for the management of acute pain after
surgical tooth extraction. Patients receiving corticosteroids in the modalities above may experience a
trivial reduction in pain intensity compared with those receiving a placebo at 6- and 24-hour follow-ups;
however, the evidence is uncertain. Future research should focus on examining the effect of cortico-
steroids on various patient-important outcomes in patients undergoing surgical tooth extractions. n

SUPPLEMENTAL DATA
Supplemental data related to this article can be found at: https://doi.org/10.1016/j.adaj.2023.04.018.

JADA 154(8) n http://jada.ada.org n August 2023 739

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
Ms. Miroshnychenko is a PhD student, Department of Health Research Dr. Polk is an assistant professor, Department of Dental Public Health,
Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, University of Pittsburgh, Pittsburgh, PA.
Canada. Dr. Moore is a professor emeritus, Department of Dental Public Health,
Ms. Azab was a bachelor of health sciences student, Faculty of Health University of Pittsburgh, Pittsburgh, PA.
Sciences, McMaster University, Hamilton, Ontario, Canada, when the work Dr. Hersh is a professor, Department of Oral Surgery and Pharmacology,
described in this article was conducted. She now is a medical student, The University of Pennsylvania, Philadelphia, PA.
Temerty Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada. Dr. Brignardello-Petersen is an assistant professor, Department of Health
Ms. Ibrahim is a master’s student, Faculty of Health Sciences, McMaster Research Methods, Evidence and Impact, McMaster University, Hamilton,
University, Hamilton, Ontario, Canada. Ontario, Canada.
Dr. Roldan is a master’s student, Department of Health Research Methods, Dr. Carrasco-Labra was a senior director, Department of Evidence
Evidence and Impact, McMaster University, Hamilton, Ontario, Canada. Synthesis and Translation Research, American Dental Association Science
Mr. Diaz Martinez is a research assistant, Department of Health Research and Research Institute, when part of the work described in this article was
Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, conducted. He now is an associate professor, Center for Integrative Global
Canada. Oral Health, School of Dental Medicine, University of Pennsylvania,
Ms. Tamilselvan is a bachelor of health sciences student, Faculty of Philadelphia, PA. Address correspondence to Dr. Carrasco-Labra, Center
Health Sciences, McMaster University, Hamilton, Ontario, Canada. for Integrative Global Oral Health, School of Dental Medicine, University of
Mr. He was a bachelor of health sciences student, Faculty of Health Pennsylvania, 240 S 40th St, 3rd Floor E, Philadelphia, PA 19104, email
Sciences, McMaster University, Hamilton, Ontario, Canada, when the work [email protected].
described in this article was conducted. He now is a medical student, The Disclosure. None of the authors reported any disclosures.
Faculty of Medicine, University of Ottawa, Ottawa, Ontario, Canada.
Ms. Urquhart was a senior research associate, Department of Evidence This project was financially supported by the US Food and Drug Adminis-
Synthesis and Translation Research, American Dental Association Science tration, Department of Health and Human Services.
and Research Institute, Chicago, IL, when part of the work described in this
article was conducted. She now is an instructor, Center for Integrative The findings and conclusions in this project are those of the authors and do not
Global Oral Health, School of Dental Medicine, University of Pennsylvania, necessarily reflect the official views, position, or policy of, nor an endorsement
Philadelphia, PA. by, the US Food and Drug Administration, US Department of Health and
Dr. Verdugo-Paiva is a chief scientific innovation officer, Epistemonikos Human Services, the US government, and the American Dental Association.
Foundation, Santiago, Chile.
Ms. Tampi is an adjunct professor, Department of Cariology, School of Presented at the International Association for Dental Research General
Dentistry, University of Michigan, Ann Arbor, MI. Session and Exhibition, June 25, 2022. https://www.iadr.org/2022iags

1. Friedman JW. The prophylactic extraction of third Framing Opioid Prescribing Guidelines for Acute Pain: submucosal, intramuscular, intravenous and per-oral
molars: a public health hazard. Am J Public Health. 2007; Developing the Evidence. National Academies Press; 2020. administration of dexamethasone on post-operative
97(9):1554-1559. 12. Rada G, Pérez D, Araya-Quintanilla F, et al.; on sequelae after mandibular impacted third molar surgery:
2. Barnes PJ. How corticosteroids control inflamma- behalf of Epistemonikos project. Epistemonikos: a a preliminary clinical comparative study. Oral Maxillofac
tion: Quintiles Prize Lecture 2005. Br J Pharmacol. 2006; comprehensive database of systematic reviews for health Surg. 2014;18(3):293-296.
148(3):245-254. decision-making. BMC Med Res Methodol. 2020;20(1): 22. Bortoluzzi MC, Capella DL, Barbieri T, Pagliarini M,
3. Ngeow WC, Lim D. Do corticosteroids still have a 286. Cavalieri M, Manfro R. A single dose of amoxicillin and
role in the management of third molar surgery? Adv Ther. 13. Third Molar L$OVE. Epistemonikos Foundation. dexamethasone for prevention of postoperative compli-
2016;33(7):1105-1139. Accessed July 9, 2021. https://app.iloveevidence.com/ cations in third molar surgery: a randomized, double-blind,
4. Markiewicz MR, Brady MF, Ding EL, Dodson TB. loves/5da4b6783c12ef714e5803f2?population¼5da4bab13 placebo controlled clinical trial. J Clin Med Res. 2013;
Corticosteroids reduce postoperative morbidity after third c12ef714e5803f4 5(1):26-33.
molar surgery: a systematic review and meta-analysis. 14. Sterne JAC, Savovic J, Page MJ, et al. RoB 2: a 23. Buyukkurt MC, Gungormus M, Kaya O. The effect
J Oral Maxillofac Surg. 2008;66(9):1881-1894. revised tool for assessing risk of bias in randomised trials. of a single dose prednisolone with and without diclofenac
5. Herrera-Briones FJ, Prados Sánchez E, Reyes BMJ. 2019;366:l4898. on pain, trismus, and swelling after removal of mandibular
Botella C, Vallecillo Capilla M. Update on the use of 15. Hultcrantz M, Rind D, Akl EA, et al. The GRADE third molars. J Oral Maxillofac Surg. 2006;64(12):1761-
corticosteroids in third molar surgery: systematic review of working group clarifies the construct of certainty of evi- 1766.
the literature. Oral Surg Oral Med Oral Pathol Oral Radiol. dence. J Clin Epidemiol. 2017;87:4-13. 24. Chugh A, Singh S, Mittal Y, Chugh V. Submucosal
2013;116(5):e342-e351. 16. Zeng L, Brignardello-Petersen R, Hultcrantz M, et al. injection of dexamethasone and methylprednisolone for
6. Moraschini V, Hidalgo R, Porto Barboza E. Effect of GRADE guidelines, 32: GRADE offers guidance on the control of postoperative sequelae after third molar
submucosal injection of dexamethasone after third molar choosing targets of GRADE certainty of evidence ratings. surgery: randomized controlled trial. Int J Oral Maxillofac
surgery: a meta-analysis of randomized controlled trials. Int J Clin Epidemiol. 2021;137:163-175. Surg. 2018;47(2):228-233.
J Oral Maxillofac Surg. 2016;45(2):232-240. 17. Afkan SS, Mofatheh M, Razavi M, Kazemi P. 25. Deo SP. Effect of submucosal injection of dexameth-
7. Larsen MK, Kofod T, Christiansen AE, Starch- Clinical evaluation of dexamethasone pills on side effects asone on post-operative sequelae of third molar surgery.
Jensen T. Different dosages of corticosteroid and routes of mandibular third molars surgery. J Res Med Dent Sci. JNMA J Nepal Med Assoc. 2011;51(182):72-78.
of administration in mandibular third molar surgery: a 2018;6(1):98-106. 26. Deo SP. Role of addition of dexamethasone to
systematic review. J Oral Maxillofac Res. 2018;9(2):e1. 18. Antunes AA, Avelar RL, Martins Neto EC, Frota R, lignocaine 2% with adrenaline in dental nerve
8. Sugragan C, Sirintawat N, Kiattavornchareon S, Dias E. Effect of two routes of administration of dexameth- blocks for third molar surgery: a prospective ran-
Khoo LK, Kumar KC, Wongsirichat N. Do corticosteroids asone on pain, edema, and trismus in impacted lower third domized control trial. Ann Maxillofac Surg. 2016;6(2):
reduce postoperative pain following third molar inter- molar surgery. Oral Maxillofac Surg. 2011;15(4):217-223. 260-266.
vention? J Dent Anesth Pain Med. 2020;20(5):281-291. 19. Atalay B, Şitilci AT, Onur ÖD. Analgesic and anti- 27. Deo SP. Single-dose of submucosal injection of
9. Guyatt GH, Oxman AD, Kunz R, Vist GE, Falck- inflammatory effects of articaine and perineural dexa- dexamethasone affects the post operative quality of life
Ytter Y, Schünemann HJ; GRADE Working Group. methasone for mandibular third molar surgery: a ran- after third molar surgery. J Maxillofac Oral Surg. 2016;
What is “quality of evidence” and why is it important to domized, double-blind study. J Oral Maxillofac Surg. 2020; 15(3):367-375.
clinicians? BMJ. 2008;336(7651):995-998. 78(4):507-514. 28. ElHag M, Coghlan K, Christmas P, Harvey W,
10. Page MJ, McKenzie JE, Bossuyt PM, et al. The 20. Bauer HC, Duarte FL, Horliana AC, et al. Assess- Harris M. The anti-inflammatory effects of dexamethasone
PRISMA 2020 statement: an updated guideline for ment of preemptive analgesia with ibuprofen coadminis- and therapeutic ultrasound in oral surgery. Br J Oral
reporting systematic reviews. BMJ. 2021;372:n71. tered or not with dexamethasone in third molar surgery: a Maxillofac Surg. 1985;23(1):17-23.
11. National Academies of Sciences Engineering Medi- randomized double-blind controlled clinical trial. Oral 29. Ghensi P, Cucchi A, Creminelli L, Tomasi C,
cine Health and Medicine Division Board on Health Care Maxillofac Surg. 2013;17(3):165-171. Zavan B, Maiorana C. Effect of oral administration of
Services Committee on Evidence-Based Clinical Practice 21. Bhargava D, Sreekumar K, Deshpande A. Effects of bromelain on postoperative discomfort after third molar
Guidelines for Prescribing Opioids for Acute Pain. intra-space injection of Twin mix versus intraoral- surgery. J Craniofac Surg. 2017;28(2):e191-e197.

740 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
30. Gholami M, Anbiaee N, Bakhshi Moqaddam Firouz 39. Larsen MK, Kofod T, Duch K. Starch-Jensen T. Ef- sodium with and without per-orally administered methyl-
Abad S, Asadi M. What are the effects of methylpred- ficacy of methylprednisolone on pain, trismus and quality prednisolone on the sequelae of impacted mandibular
nisolone injection into the masseter and gluteal muscle on of life following surgical removal of mandibular third third molar removal: a cohort randomized double-blind
pain, edema and trismus after impacted lower third molar molars: a double-blind, split-mouth, randomised clinical trial. Indian J Dent. 2016;7(1):11-16.
surgery? A randomized clinical trial. J Oral Maxillofac Surg. controlled trial. Med Oral Patol Oral Cir Bucal. 2021;26(2): 49. Sahu S, Patley A, Kharsan V, Madan RS,
2021;79(9):1829-1836. e156-e163. Manjula V, Tiwari RVC. Comparative evaluation of ef-
31. Gopinath KA, Chakraborty M, Arun V. Compara- 40. Lim D, Ngeow WC. A comparative study on the ficacy and latency of twin mix vs 2% lignocaine HCL with
tive evaluation of submucosal and intravenous dexa- efficacy of submucosal injection of dexamethasone versus 1:80000 epinephrine in surgical removal of impacted
methasone on postoperative sequelae following third methylprednisolone in reducing postoperative sequelae mandibular third molar. J Family Med Prim Care. 2020;
molar surgery: a prospective randomized control study. Int after third molar surgery. J Oral Maxillofac Surg. 2017; 9(2):904-908.
J Oral Care Res. 2017;5(3):191-195. 75(11):2278-2286. 50. Saravanan K, Kannan R, John RR, Nantha
32. Gozali P, Boonsiriseth K, Kiattavornchareon S, 41. Majid OW. Submucosal dexamethasone injection Kumar C. A single pre operative dose of sub mucosal
Khanijou M, Wongsirichat N. Decreased post-operative improves quality of life measures after third molar surgery: dexamethasone is effective in improving post operative
pain using a sublingual injection of dexamethasone (8 a comparative study. J Oral Maxillofac Surg. 2011;69(9): quality of life in the surgical management of impacted
mg) in lower third molar surgery. J Dent Anesth Pain Med. 2289-2297. third molars: a comparative randomised prospective study.
2017;17(1):47-53. 42. Majid OW, Mahmood WK. Effect of submucosal and J Maxillofac Oral Surg. 2016;15(1):67-71.
33. Grossi GB, Maiorana C, Garramone RA, et al. Effect intramuscular dexamethasone on postoperative sequelae 51. Schmelzeisen R, Frölich JC. Prevention of post-
of submucosal injection of dexamethasone on post- after third molar surgery: comparative study. Br J Oral operative swelling and pain by dexamethasone after
operative discomfort after third molar surgery: a pro- Maxillofac Surg. 2011;49(8):647-652. operative removal of impacted third molar teeth. Eur J
spective study. J Oral Maxillofac Surg. 2007;65(11):2218- 43. Majid OW, Mahmood WK. Use of dexamethasone Clin Pharmacol. 1993;44(3):275-277.
2226. to minimise post-operative sequelae after third molar sur- 
52. Selimovic E, Ibrahimagic-Seper  sic I, Sivic S,
L, Si
34. Ibikunle AA, Adeyemo WL, Ladeinde AL. Effect of gery: comparison of five different routes of administration. Huseinagic S. Prevention of trismus with different phar-
submucosal or oral administration of prednisolone on Oral Surg. 2013;6(4):200-208. macological therapies after surgical extraction of impacted
postoperative sequelae following surgical extraction of 44. Mojsa IM, Pokrowiecki R, Lipczynski K, mandibular third molar. Med Glas (Zenica). 2017;14(1):
impacted mandibular third molar: a randomized controlled Czerwonka D, Szczeklik K, Zaleska M. Effect of sub- 145-151.
study. Niger Med J. 2016;57(5):272-279. mucosal dexamethasone injection on postoperative 53. Simone JL, Jorge WA, Horliana AC, Canaval TG,
35. Imon AA, Akhter M, Costa PD, Sabrina F, pain, oedema, and trismus following mandibular third Tortamano IP. Comparative analysis of preemptive
Haque N. Comparative study for effectiveness of oral molar surgery: a prospective, randomized, double-blind analgesic effect of dexamethasone and diclofenac
dexamethasone for assessment of pain, swelling and clinical trial. Int J Oral Maxillofac Surg. 2017;46(4): following third molar surgery. Braz Oral Res. 2013;27(3):
trismus following impacted 3rd molar surgery. SVOA 524-530. 266-271.
Dentistry. 2021;2(1):20-24. 45. Nair RB, Rahman NM, Ummar M. Effect of sub- 54. Tiigimae-Saar J, Leibur E, Tamme T. The effect
36. Kang SH, Choi YS, Byun IY, Kim MK. Effect of mucosal injection of dexamethasone on postoperative of prednisolone on reduction of complaints after
preoperative prednisolone on clinical postoperative discomfort after third molar surgery: a prospective study. J impacted third molar removal. Stomatologija. 2010;
symptoms after surgical extractions of mandibular third Contemp Dent Pract. 2013;14(3):401-404. 12(1):17-22.
molars. Aust Dent J. 2010;55(4):462-467. 46. Pansard HB, Prado MC, Marchi GF, Sfreddo CS, 55. Vegas-Bustamante E, Micó-Llorens J, Gargallo-
37. Khalida B, Fazal M, Muntaha ST, Khan K. Effect of Skupien JA. The impact of prior use of corticosteroid to Albiol J, Satorres-Nieto M, Berini-Aytés L, Gay-
submucosal injection of dexamethasone on post- dental extraction on oral health-related quality-of-life and Escoda C. Efficacy of methylprednisolone injected into
operative swelling and trismus following impacted clinical outcomes: a randomized clinical trial. J Oral the masseter muscle following the surgical extraction of
mandibular third molar surgery. Pak Oral Dent J. 2017; Maxillofac Surg. 2020;78(12):2153.e1-e53.e9. impacted lower third molars. Int J Oral Maxillofac Surg.
37(2):231-234. 47. Pedersen A. Decadronphosphate in the relief of 2008;37(3):260-263.
38. Larsen MK, Kofod T, Duch K. Starch-Jensen T. complaints after third molar surgery: a double-blind, 56. Zerener T, Aydintug YS, Sencimen M, et al. Clinical
Short-term haematological parameters following surgical controlled trial with bilateral oral surgery. Int J Oral comparison of submucosal injection of dexamethasone
removal of mandibular third molars with different doses of Surg. 1985;14(3):235-240. and triamcinolone acetonide on postoperative discomfort
methylprednisolone compared with placebo: a randomized 48. Prashar DV, Pahwa D, Kalia V, Jindal G, Kaur R. after third molar surgery. Quintessence Int. 2015;46(4):
controlled trial. J Oral Maxillofac Res. 2020;11(2):e3. A comparative evaluation of the effect of diclofenac 317-326.

JADA 154(8) n http://jada.ada.org n August 2023 741

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 APPENDIX

Search strategy
The following strategy was used to search in Epistemonikos database. We adapted it to the syntax
for other databases: (((((third* OR 3rd) AND molar*) OR (wisdom* AND (tooth* OR teeth*)))))
AND (corticostero* OR corticoid* OR steroid OR steroids OR glucocortico* OR beclometasone*
OR budesonide* OR betamethasone* OR fluticasone* OR triamcinolone* OR ciclesonide* OR
dexamet* OR dextenza* OR methylprednisolone* OR prednisolone* OR prednisone* OR hy-
drocortisone* OR cortisone* OR mineralocortico* OR fludrocortisone*).

Risk difference Risk difference

Corticosteroid Placebo Weight, Mantel-Haenszel, Mantel-Haenszel,
Study or subgroup Events Total Events Total % random, 95% CI random, 95% CI

ElHag and Colleagues,28 1985 0 38 0 32 3.3 0.00 (–0.05 to 0.05)

Pedersen and Colleagues,47 1985 0 30 0 30 2.5 0.00 (–0.06 to 0.06)
Schmelzeisen and Colleagues,51 1993 5 25 3 25 0.2 0.08 (–0.12 to 0.28)
Grossi and Colleagues,33 2007A 0 18 0 11 0.5 0.00 (–0.13 to 0.13)
Grossi and Colleagues,33 2007B 0 20 0 12 0.6 0.00 (–0.12 to 0.12)
Vegas-Bustamante and Colleagues,55 2008 0 35 0 35 3.4 0.00 (–0.05 to 0.05)
Tiigimae and Colleagues,54 2010 0 38 0 40 4.2 0.00 (–0.05 to 0.05)
Majid and Colleagues,41 2011A 0 11 0 5 0.2 0.00 (–0.25 to 0.25)
Majid and Colleagues,41 2011B 0 11 0 6 0.2 0.00 (–0.22 to 0.22)
Majid and Colleagues,42 2011A 0 10 0 5 0.2 0.00 (–0.25 to 0.25)
Majid and Colleagues,42 2011B 0 10 0 5 0.2 0.00 (–0.25 to 0.25)
Bortoluzzi and Colleagues,22 2013A 0 12 0 12 0.5 0.00 (–0.15 to 0.15)
Bortoluzzi and Colleagues,22 2013B 1 14 0 12 0.3 0.07 (–0.11 to 0.26)
Majid and Colleagues,43 2013A 0 12 0 4 0.1 0.00 (–0.28 to 0.28)
Majid and Colleagues,43 2013B 0 11 0 4 0.1 0.00 (–0.29 to 0.29)
Majid and Colleagues,43 2013C 0 12 0 4 0.1 0.00 (–0.28 to 0.28)
Nair and Colleagues,45 2013 0 50 0 50 6.7 0.00 (–0.04 to 0.04)
Ghensi and Colleagues,29 2017A 0 20 0 21 1.2 0.00 (–0.09 to 0.09)
Ghensi and Colleagues,29 2017B 0 20 0 19 1.1 0.00 (–0.09 to 0.09)
Gopinath and Colleagues,31 2017 0 40 0 40 4.4 0.00 (–0.05 to 0.05)
Khalida and Colleagues,37 2017 0 30 0 30 2.5 0.00 (–0.06 to 0.06)
Lim and Colleagues,40 2017A 0 20 0 10 0.5 0.00 (–0.14 to 0.14)
Lim and Colleagues,40 2017B 0 20 0 10 0.5 0.00 (–0.14 to 0.14)
Mojsa and Colleagues,44 2017A 0 30 0 15 1.1 0.00 (–0.10 to 0.10)
Mojsa and Colleagues,44 2017B 0 30 0 15 1.1 0.00 (–0.10 to 0.10)
Chugh and Colleagues,24 2018A 0 23 0 9 0.5 0.00 (–0.15 to 0.15)
Chugh and Colleagues,24 2018B 0 20 0 8 0.4 0.00 (–0.16 to 0.16)
Atalay and Colleagues,19 2020A 0 20 0 10 0.5 0.00 (–0.14 to 0.14)
Atalay and Colleagues,19 2020B 0 20 0 10 0.5 0.00 (–0.14 to 0.14)
Larsen and Colleagues,38 2020A 5 26 2 9 0.1 –0.03 (–0.34 to 0.28)
Larsen and Colleagues,38 2020B 5 26 2 9 0.1 –0.03 (–0.34 to 0.28)
Larsen and Colleagues,38 2020C 6 26 2 8 0.1 –0.02 (–0.36 to 0.32)
Pansard and Colleagues,46 2020 0 44 0 49 5.8 0.00 (–0.04 to 0.04)
Imon and Colleagues,35 2021 0 147 0 147 56.4 0.00 (–0.01 to 0.01)

Total (95% CI) 919 711 100.0 0.00 (-0.01 to 0.01)

Total events 22 9
Heterogeneity: τ2 = 0.00, χ 233 = 1.43 (P = 1.00), I 2 = 0% –1 –0.5 0 0.5 1
Test for overall effect: z = 0.06 (P = .95) Favors corticosteroid Favors placebo

eFigure 1. Corticosteroids vs a placebo for the outcome of postoperative infection (adverse effect). A, B, and C represent different interventions within
the same study.

741.e1 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Risk difference, Risk difference,
Corticosteroid Placebo Weight, Mantel-Haenszel, Mantel-Haenszel,
Study or subgroup Events Total Events Total % random, 95% CI random, 95% CI

ElHag and Colleagues,28 1985 2 38 0 32 14.9 0.05 (–0.03 to 0.14)

Pedersen and Colleagues,47 1985 0 30 0 30 28.9 0.00 (–0.06 to 0.06)
Majid and Colleagues,41 2011A 0 11 0 5 1.8 0.00 (–0.25 to 0.25)
Majid and Colleagues,41 2011B 0 11 0 6 2.3 0.00 (–0.22 to 0.22)
Majid and Colleagues,42 2011A 0 10 0 5 1.8 0.00 (–0.25 to 0.25)
Majid and Colleagues,42 2011B 0 10 0 5 1.8 0.00 (–0.25 to 0.25)
Bortoluzzi and Colleagues,22 2013A 0 12 1 12 2.8 –0.08 (–0.29 to 0.12)
Bortoluzzi and Colleagues,22 2013B 0 14 1 12 3.0 –0.08 (–0.28 to 0.11)
Majid and Colleagues,43 2013A 0 12 0 4 1.4 0.00 (–0.28 to 0.28)
Majid and Colleagues,43 2013B 0 11 0 4 1.4 0.00 (–0.29 to 0.29)
Majid and Colleagues,43 2013C 0 12 0 4 1.4 0.00 (–0.28 to 0.28)
Khalida and Colleagues,37 2017 0 30 0 30 28.9 0.00 (–0.06 to 0.06)
Chugh and Colleagues,24 2018A 0 23 0 9 5.3 0.00 (–0.15 to 0.15)
Chugh and Colleagues,24 2018B 0 20 0 8 4.3 0.00 (–0.16 to 0.16)
Total (95% CI) 244 166 100 0.00 (–0.03 to 0.04)
Total events
Heterogeneity: τ2 = 0.00, χ 213 = 2.77 (P = 1.00), I 2 = 0%
Test for overall effect: z = 0.18 (P = .86) –1 –0.5 0 0.5 1
Favors corticosteroid Favors placebo

eFigure 2. Corticosteroids vs a placebo for the outcome of alveolar osteitis (adverse effect). A, B, and C represent different interventions within the same
study.

Risk difference Risk difference

Corticosteroid Placebo Weight, Mantel-Haenszel, Mantel-Haenszel,
Study or subgroup Events Total Events Total % random, 95% CI random, 95% CI

Schmelzeisen and Colleagues,51 1993 4 25 5 25 33.3 –0.04 (–0.25 to 0.17)

Deo and Colleagues,27 2016 1 19 7 11 29.7 –0.58 (–0.89 to –0.28)
Selimovic and Colleagues,52 2017 0 20 0 20 37.0 0.00 (–0.09 to 0.09)
Total (95% CI) 64 56 100.0 –0.19 (–0.54 to 0.16)
Total events 5 12
Heterogeneity: τ2 = 0.08, χ 22 = 19.74 (P < .0001), I 2 = 90%
–1 –0.5 0 0.5 1
Test for overall effect: z = 1.04 (P = .30)
Favors corticosteroid Favors placebo

eFigure 3. Corticosteroids vs a placebo for the outcome of nausea or vomiting adverse effect.

Risk difference Risk difference

Corticosteroid Placebo Weight, Mantel-Haenszel, Mantel-Haenszel,
Study or subgroup Events Total Events Total % random, 95% CI random, 95% CI

1.2.1 Dexamethasone
Schmelzeisen and Colleagues,51 1993 4 25 5 25 33.3 –0.04 (–0.25 to 0.17)
Deo and Colleagues,27 2016 1 19 7 11 29.7 –0.58 (–0.89 to –0.28)
Subtotal (95% CI) 44 36 63.0 –0.30 (–0.83 to 0.23)
Total events 5 12
Heterogeneity: τ2 = 0.13, χ 21 = 8.37 (P = .004), I 2 = 88%
Test for overall effect: z = 1.11 (P = .27)
1.2.2 Methylprednisolone
Selimovic and Colleagues,52 2017 0 20 0 20 37.0 0.00 (–0.09 to 0.09)
Subtotal (95% CI) 20 20 37.0 0.00 (–0.09 to 0.09)
Total events 0 0
Heterogeneity: Not applicable
Test for overall effect: z = 0.00 (P = 1.00)

Total (95% CI) 64 56 100.0 –0.19 (–0.54 to 0.16)

Total events 5 12
Heterogeneity: τ2 = 0.08, χ 22 = 19.74 (P < .0001), I 2 = 90%
–1 –0.5 0 0.5 1
Test for overall effect: z = 1.04 (P = .30)
Favors corticosteroid Favors placebo
Test for subgroup differences: χ 21 = 1.19, (P = .28), I 2 = 15.9%

eFigure 4. Gastrointestinal adverse effects of nausea or vomiting (corticosteroid type).

JADA 154(8) n http://jada.ada.org n August 2023 741.e2

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 Risk difference Risk difference
Corticosteroid Placebo Weight, Mantel-Haenszel, Mantel-Haenszel,
Study or subgroup Events Total Events Total % random, 95% CI random, 95% CI

1.3.1 Multi-dose (oral)

Schmelzeisen and Colleagues,51 1993 4 25 5 25 33.3 –0.04 (–0.25 to 0.17)
Selimovic and Colleagues,52 2017 0 20 0 20 37.0 0.00 (–0.09 to 0.09)
Subtotal (95% CI) 45 45 70.3 –0.01 (–0.09 to 0.08)
Total events 4 5
Heterogeneity: τ2 = 0.00, χ 21 = 0.25 (P = .62), I 2 = 0%
Test for overall effect: z = 0.15 (P = .88)
1.3.2 Single dose (submucosal)
Deo and Colleagues,27 2016 1 19 7 11 29.7 –0.58 (–0.89 to –0.28)
Subtotal (95% CI) 19 11 29.7 –0.58 (–0.89 to –0.28)
Total events 1 7
Heterogeneity: Not applicable
Test for overall effect: z = 3.79 (P = .0001)

Total (95% CI) 64 56 100.0 –0.19 (–0.54 to 0.16)

Total events 5 12
Heterogeneity: τ2 = 0.08, χ 22 = 19.74 (P < .0001), I 2 = 90%
–1 –0.5 0 0.5 1
Test for overall effect: z = 1.04 (P = .30)
Favors corticosteroid Favors placebo
Test for subgroup differences: χ 21 = 13.06, (P = .0003), I 2 = 92.3%

eFigure 5. Gastrointestinal adverse effects of nausea or vomiting (route of administration).

741.e3 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
eTable 1. Preferred Reporting Items for Systematic Reviews and Meta-Analysis checklist.

SECTION AND TOPIC ITEM NO. CHECKLIST ITEM LOCATION WHERE ITEM IS REPORTED
Title

Title 1 Identify the report as a systematic review. Paragraph 1

Abstract

Abstract 2 See the Preferred Reporting Items for Systematic Reviews and Meta- Paragraph 1
Analysis 2020 for abstracts checklist.

Introduction

Rationale 3 Describe the rationale for the review in the context of existing knowledge. Paragraph 4

Objectives 4 Provide an explicit statement of the objectives or questions the review Paragraph 4
addresses.

Methods

Eligibility criteria 5 Specify the inclusion and exclusion criteria for the review and how studies Paragraph 5
were grouped for the syntheses.

Information sources 6 Specify all databases, registers, websites, organizations, reference lists, Paragraph 5
and other sources searched or consulted to identify studies. Specify the
date when each source was last searched or consulted.

Search strategy 7 Present the full search strategies for all databases, registers, and websites, Paragraph 5, search strategy section in Appendix
including any filters and limits used.

Selection process 8 Specify the methods used to decide whether a study met the inclusion Paragraph 6
criteria of the review, including how many reviewers screened each
record and each report retrieved, whether they worked independently,
and, if applicable, details of automation tools used in the process.

Data collection process 9 Specify the methods used to collect data from reports, including how Paragraph 7
many reviewers collected data from each report, whether they worked
independently, any processes for obtaining or confirming data from study
investigators, and, if applicable, details of automation tools used in the
process.

Data items 10a List and define all outcomes for which data were sought. Specify whether Paragraph 7
all results that were compatible with each outcome domain in each study
were sought (for example, for all measures, time points, and analyses),
and if not, the methods used to decide which results to collect.
10b List and define all other variables for which data were sought (for Paragraph 7-8
example, participant and intervention characteristics, funding sources).
Describe any assumptions made about any missing or unclear
information.

Study risk of bias assessment 11 Specify the methods used to assess risk of bias in the included studies, Paragraph 7
including details of the tools used, how many reviewers assessed each
study, whether they worked independently, and, if applicable, details of
automation tools used in the process.
Effect measures 12 Specify for each outcome the effect measures (for example, risk ratio and Paragraph 7-8
mean difference) used in the synthesis or presentation of results.

Synthesis methods 13a Describe the processes used to decide which studies were eligible for Paragraph 6-7
each synthesis (for example, tabulating the study intervention
characteristics and comparing against the planned groups for each
synthesis (item no. 5).
13b Describe any methods required to prepare the data for presentation or Paragraph 7-8
synthesis, such as handling of missing summary statistics or data
conversions.

13c Describe any methods used to tabulate or visually display results of Paragraph 7-8
individual studies and syntheses.
13d Describe any methods used to synthesize results and provide a rationale Paragraph 7-8
for the choices. If meta-analysis was performed, describe the models,
methods to identify the presence and extent of statistical heterogeneity,
and software packages used.

13e Describe any methods used to explore possible causes of heterogeneity Paragraph 9
among study results (for example, subgroup analysis and meta-
regression).
13f Describe any sensitivity analyses conducted to assess robustness of the NA*
synthesized results.

* NA: Not applicable.

JADA 154(8) n http://jada.ada.org n August 2023 741.e4

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
eTable 1. Continued

SECTION AND TOPIC ITEM NO. CHECKLIST ITEM LOCATION WHERE ITEM IS REPORTED
Reporting bias assessment 14 Describe any methods used to assess risk of bias due to missing results in NA
a synthesis (arising from reporting biases).

Certainty assessment 15 Describe any methods used to assess certainty (or confidence) in the body Paragraph 8
of evidence for an outcome.
Results

Study selection 16a Describe the results of the search and selection process, from the number Paragraph 9
of records identified in the search to the number of studies included in
the review, ideally using a flow diagram.
16b Cite studies that might appear to meet the inclusion criteria, but which NA
were excluded, and explain why they were excluded.

Study characteristics 17 Cite each included study and present its characteristics. Paragraph 9-10

Risk of bias in studies 18 Present assessments of risk of bias for each included study. Paragraph 10, eTable 2

Results of individual studies 19 For all outcomes, present, for each study: (a) summary statistics for each NA
group (where appropriate) and (b) an effect estimate and its precision (for
example, CI/credible interval), ideally using structured tables or plots.

Results of syntheses 20a For each synthesis, briefly summarize the characteristics and risk of bias Paragraph 10-12
among contributing studies.

20b Present results of all statistical syntheses conducted. If meta-analysis was Paragraph 10-12
done, present for each the summary estimate and its precision (for
example, CI/credible interval) and measures of statistical heterogeneity. If
comparing groups, describe the direction of the effect.

20c Present results of all investigations of possible causes of heterogeneity Figures, eFigure 4, and eFigure 5
among study results.

20d Present results of all sensitivity analyses conducted to assess the NA

robustness of the synthesized results.

Reporting biases 21 Present assessments of risk of bias due to missing results (arising from NA
reporting biases) for each synthesis assessed.

Certainty of evidence 22 Present assessments of certainty (or confidence) in the body of evidence Table 2
for each outcome assessed.

Discussion

Discussion 23a Provide a general interpretation of the results in the context of other Paragraph 12-13
evidence.

23b Discuss any limitations of the evidence included in the review. Paragraph 12-13

23c Discuss any limitations of the review processes used. Paragraph 15-16

23d Discuss implications of the results for practice, policy, and future research. Paragraph 16

Other Information

Registration and protocol 24a Provide registration information for the review, including register name NA
and registration number, or state that the review was not registered.

24b Indicate where the review protocol can be accessed, or state that a 4
protocol was not prepared.

24c Describe and explain any amendments to information provided at NA

registration or in the protocol.

Support 25 Describe sources of financial or nonfinancial support for the review, and Article footnote
the role of the funders or sponsors in the review.

Competing interests 26 Declare any competing interests of review authors. Article footnote

Availability of data, code, and 27 Report which of the following are publicly available and where they can NA
other materials be found: template data collection forms; data extracted from included
studies; data used for all analyses; analytic code; any other materials used
in the review.

741.e5 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
eTable 2. Risk of bias assessment.

DEVIATIONS FROM THE MISSING SELECTION OF

INTENDED OUTCOME MEASUREMENT THE REPORTED
STUDY* RANDOMIZATION INTERVENTION DATA OF OUTCOME RESULTS
Pain at 6 h

Buyukkurt and Probably low Low Low Low Low

colleagues,23 2006

Vegas-Bustamante Low Low Probably low Low High

and colleagues,55
2008

Bauer and Low Low Probably low Low Low

colleagues, 201220

Bortoluzzi and Low Low Low Low High

colleagues,22
2013A

Bortoluzzi and Low Low Low Low High

colleagues,22 2013B

Simone and Low Low High Low High

colleagues,53 2013

Mojsa and Low Low Probably high Low High

colleagues,44
2017A

Mojsa and Low Low Probably high Low High

colleagues,44 2017B

Afkan and Probably low Low High Low Probably low

colleagues,17
2018A

Afkan and Probably low Low High Low Probably low

colleagues,17 2018B

Pain at 24 h

Vegas-Bustamante Low Low Probably low Low High

and colleagues,55
2008

Kang and Probably high High High High Low

colleagues,36
2010A

Kang and Probably high High High High Low

colleagues,36 2010B

Antunes and High High Probably low Probably low Low

colleagues,18
2011A

Antunes and High High Probably low Probably low Low

colleagues,18 2011B

Deo and Low Low Low Low Low

colleagues,25 2011

Majid and Probably high High Low Probably low Low

colleagues,41
2011A

Majid and Probably high High Low Probably low Low

colleagues,41 2011B

Majid and Probably low High Low High Low

colleagues,42
2011A

Majid and Probably low High Low High Low

colleagues,42 2011B

Bauer and Low Low Probably low Low Low

colleagues,20 2012

* A, B, and C represent different interventions within the same study.

JADA 154(8) n http://jada.ada.org n August 2023 741.e6

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 eTable 2. Continued

DEVIATIONS FROM THE MISSING SELECTION OF

INTENDED OUTCOME MEASUREMENT THE REPORTED
STUDY* RANDOMIZATION INTERVENTION DATA OF OUTCOME RESULTS
Bortoluzzi and Low Low Low Low High
colleagues,22
2013A

Bortoluzzi and Low Low Low Low High

colleagues,22 2013B

Majid and Probably high Probably high Probably low High Low
colleagues,43
2013A

Majid and Probably high Probably high Probably low High Low
colleagues,43 2013B

Majid and Probably high Probably high Probably low High Low
colleagues,43
2013C

Simone and Low Low High Low High

colleagues,53 2013

Bhargava and Probably low Low Low Low Probably high

colleagues,21
2014A

Bhargava and Probably low Low Low Low Probably high

colleagues,21 2014B

Bhargava and Probably low Low Low Low Probably high

colleagues,21
2014C

Zerener and Probably high Probably high Low High Probably low
colleagues,56 2015B

Deo and Probably low Low High Low Low

colleagues,26 2016

Ibikunle and Probably high Probably high Low High Low

colleagues,34
2016A

Ibikunle and Probably high Probably high Low High Low

colleagues,34 2016B

Prashar and Probably low Low Low Low Probably low

colleagues,48 2016

Saravanan and Probably high Probably high Low High Probably low
colleagues,50
2016A

Saravanan and Probably high Probably high Low High Probably low
colleagues,50 2016B

Gozali and Probably low Low Probably low Low Probably low
colleagues,32 2017

Lim and Low Probably low Low Low High

colleagues,40
2017A

Mojsa and Low Low Probably high Low High

colleagues,44
2017A

Mojsa and Low Low Probably high Low High

colleagues,44 2017B

Chugh and Low Probably high Probably high High Low

colleagues,24
2018A

Sahu and Probably low Probably low Low Low Probably low
colleagues,49 2020

741.e7 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
eTable 2. Continued

DEVIATIONS FROM THE MISSING SELECTION OF

INTENDED OUTCOME MEASUREMENT THE REPORTED
STUDY* RANDOMIZATION INTERVENTION DATA OF OUTCOME RESULTS
Gholami and Probably low Probably high Low High Low
colleagues,30
2021A

Gholami and Probably low Probably high Low High Low

colleagues,30 2021B

Larsen and Low Low Low Low Probably low

colleagues,39
2021A

Larsen and Low Low Low Low Probably low

colleagues,39 2021B

Larsen and Low Low Low Low Probably low

colleagues,39
2021C

Adverse Effects
Postoperative
infection
ElHag and Low High Low Low Low
colleagues,28
1985
Pedersen and Probably low Low Low Low Probably low
colleagues,47
1985
Schmelzeisen and Probably low Probably high High Probably high Probably low
colleagues,51
1993
Grossi and Probably low High Low Low Probably low
colleagues,33
2007A
Grossi and Probably low High Low Low Probably low
colleagues,33
2007B
Vegas- Low Low Probably low Low Probably low
Bustamante and
colleagues,55
2008
Tiigimae and Probably high High Low High Low
colleagues,54
2010
Majid and Probably high High Low Low Probably low
colleagues,41
2011A
Majid and Probably high High Low Low Probably low
colleagues,41
2011B
Majid and Probably low High Low Probably high Probably low
colleagues,42
2011A
Majid and Probably low High Low Probably high Probably low
colleagues,42
2011B
Bortoluzzi and Low Low Low Low Low
colleagues,22
2013A
Bortoluzzi and Low Low Low Low Low
colleagues,22
2013B
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013A

JADA 154(8) n http://jada.ada.org n August 2023 741.e8

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
 eTable 2. Continued

DEVIATIONS FROM THE MISSING SELECTION OF

INTENDED OUTCOME MEASUREMENT THE REPORTED
STUDY* RANDOMIZATION INTERVENTION DATA OF OUTCOME RESULTS
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013B
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013C
Nair and Probably high Probably high Low Low Low
colleagues,45
2013
Ghensi and Probably low Probably high Low Low Low
colleagues,29
2017A
Ghensi and Probably low Probably high Low Low Low
colleagues,29
2017B
Gopinath and Probably high Probably high Probably low Low Low
colleagues,31
2017
Khalida and Probably high Probably high Low Low Low
colleagues,37
2017
Lim and Low Probably low Low Low Low
colleagues,40
2017A
Mojsa and Low Low Low Low Probably low
colleagues,44
2017A
Mojsa and Low Low Low Low Probably low
colleagues,44
2017B
Chugh and Low Probably high Low Low Probably high
colleagues,24
2018A
Atalay and Probably low Low Low Low Probably low
colleagues,19
2020A
Atalay and Probably low Low Low Low Probably low
colleagues,19
2020B
Pansard and Low Low High Low Probably low
colleagues,46
2020
Larsen and Low Low Low Low High
colleagues,38
2020A
Larsen and Low Low Low Low High
colleagues,38
2020B
Larsen and Low Low Low Low High
colleagues,38
2020C
Imon and Probably high Probably high Low High Probably low
colleagues,35
2021
Alveolar osteitis
ElHag and Low High Low Low Low
colleagues,28
1985
Pedersen and Probably low Low Low Low Probably low
colleagues,47
1985

741.e9 JADA 154(8) n http://jada.ada.org n August 2023

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.
eTable 2. Continued

DEVIATIONS FROM THE MISSING SELECTION OF

INTENDED OUTCOME MEASUREMENT THE REPORTED
STUDY* RANDOMIZATION INTERVENTION DATA OF OUTCOME RESULTS
Majid and Probably high High Low Low Probably low
colleagues,41
2011A
Majid and Probably high High Low Low Probably low
colleagues,41
2011B
Majid and Probably low High Low Probably high Probably low
colleagues,42
2011A
Majid and Probably low High Low Probably high Probably low
colleagues,42
2011B
Bortoluzzi and Low Low Low Low Low
colleagues,22
2013A
Bortoluzzi and Low Low Low Low Low
colleagues,22
2013B
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013A
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013B
Majid and Probably high Probably high Probably low Low Low
colleagues,43
2013C
Khalida and Probably high Probably high Low Low Low
colleagues,37
2017
Chugh and Low Probably high Low Low Probably high
colleagues,24
2018A
Nausea or
vomiting
Schmelzeisen and Probably low Probably high High Probably high Low
colleagues, 1993
Deo and Probably low Low High Low Low
colleagues,27
2016

Selimovic and Probably high Probably high Low High High

colleagues,52
2017

JADA 154(8) n http://jada.ada.org n August 2023 741.e10

Downloaded for Anonymous User (n/a) at University of the Andes from ClinicalKey.com by Elsevier on November 02,
2024. For personal use only. No other uses without permission. Copyright ©2024. Elsevier Inc. All rights reserved.