Keywords
photodynamic therapy - orthodontics - antimicrobial agent - inflammation - gum
Introduction
Maintaining adequate oral hygiene represents a major challenge[1]
[2]
[3] for patients undergoing orthodontic treatment with fixed appliances.[4]
[5]
Methods for calculus removal by the dentist, during the oral prophylaxis procedure,
are performed using periodontal curettes and ultrasound (US). There are other methods
that can also help with microbiological control[6]
[7] in the oral cavity, such as antimicrobial mouthwashes, especially those based on
chlorhexidine (CHX).[8]
[9] However, some side effects have been demonstrated with the long-term use of CHX,
such as changes in the color of teeth and tongue, changes in taste, burning sensation,
and genotoxicity of oral epithelial cells.[10]
At the same time, in recent years, methods such as laser irradiation,[11] and photodynamic therapy (PDT), which is associated with a photosensitizer (PS),
have been increasingly used in microbiological and inflammatory control.[12] PS, when applied, accumulates intensely in the tissues. PS molecules will absorb
light of the appropriate wavelength, initiating an activation process that leads cells
to selective destruction. These cells present selective accumulation of PS, and with
exposure to light, they are damaged by the phototoxic and oxidative reaction.[13]
[14] The main PSs used in dentistry are methylene blue and toluidine blue.[8]
[15]
Although the use of diode lasers is on the rise for periodontal treatments, the difficulty
in standardizing oral hygiene and individual differences brings uncertainty in the
results.[11] The effects of PDT on patients using orthodontic braces have demonstrated antimicrobial
effects, reducing the risk of infections, inflammation, periodontal diseases, and
cavities.[5]
[12]
[14]
[15]
Considering the importance of evaluating the use of PDT in microbiological and inflammatory
control in patients during orthodontic treatment,[5]
[12]
[15] and the absence of network systematic reviews (network meta-analysis [NMA]) to analyze
different treatment combinations, the objective of this systematic review and the
NMA was to determine the effect of PDT on the reduction of gingival inflammation evaluated
in clinical trials, measured using different indices, periodontal, microbiological
culture, and inflammatory cytokines.
Methods
Focus Question
This systematic review was performed to answer the following clinical question: Is
there any scientific evidence of the effectiveness of PDT in controlling gingival
health in orthodontic patients compared with patients who received non–PDT-based treatment?
PICO question and eligibility criteria are detailed in [Table 1].
Table 1
Criteria (PICOS, inclusion and exclusion) for study selection and research strategy
|
PICOS
|
|
Participant (P)
|
Orthodontic patients aged at least 13 years
|
|
Intervention (I)
|
Photodynamic therapy (PDT)
|
|
Comparison (C)
|
Subjected to non–PDT-based treatment
|
|
Result (O)
|
Primary outcome:
|
Decreased gingival inflammation (PD, PI, GBI, GI)
|
|
Periodontal pocket probing depth (PD)
|
|
Decrease in dental Plaque Index (PI)
|
|
Decrease in the Gingival Bleeding Index (GBI)
|
|
Decrease in the Gingival Index (GI)
|
|
Secondary outcome:
|
Assessment of oral microbiological culture (OMC)
|
|
Biochemical assessment of inflammation in gingival crevicular fluid (GCF)
|
|
Study (S)
|
Randomized and non-randomized clinical trials (RCT and NRCT)
|
|
Criteria
|
|
Inclusion
|
Articles that evaluated the influence of PDT on pathogenic microorganisms. Report
the gingival and periodontal condition, minimum of 3 months of fixed orthodontic treatment
in both arches, and not having received anti-inflammatory and antimicrobial therapy
in the last 30 days.
|
|
Exclusion
|
Patients using systemic drugs and systemic disease, have not received oral hygiene
instructions and dental cleaning prior to clinical trial, use of oral rinse over the
past 30 days, orthodontic retreatment, smokers, pregnant women, erupting permanent
teeth, control studies without group or without microbial evaluation. Case reports,
case series, animal studies, in vitro studies, editor letters, literature review and
editorials.
|
|
Database
|
Research strategy
|
|
Pubmed (Medline)
|
((orthodontics [MeSH Terms] OR orthodontic [Title/Abstract] OR "orthodontic patients"
[Title/Abstract] OR "orthodontic treatment" [Title/Abstract] OR "periodontal index"
[MeSH Terms] OR "periodontal diseases" [MeSH Terms] OR periodontitis [MeSH Terms]
OR periodontics [MeSH Terms] OR "periodontal patients" [Title/Abstract] OR periodontal
[Title/Abstract] OR "periodontal treatment" [Title/Abstract] OR gingivitis [MeSH Terms]
OR gingiva [MeSH Terms] OR "periodontitis patients" [Title/Abstract] OR caries [MeSH
Terms] OR “carious dentin” [MeSH Terms] OR “caries disease” [MeSH Terms] OR “streptococcus
mutans” [MeSH Terms] OR streptococcus [MeSH Terms]) AND (photochemotherapy [MeSH Terms]
OR "photodynamic therapy” [Title/Abstract]) AND ("anti-infective agents" [MeSH Terms]
OR antimicrobial [Title/Abstract] OR bactericidal [Title/Abstract] OR bacteriostatic
[Title/Abstract] OR "anti-inflammatory agents" [MeSH Terms] OR "antiviral agents"
[MeSH Teraams] OR "antifungal agents" [MeSH Terms] OR “cariogenic biofilm” [Title/Abstract]
OR biofilms [MeSH Terms] OR apoptosis [MeSH Terms] OR efficacy [Title/Abstract] OR
"side effects" [Title/Abstract] OR "adverse effects" [MeSH Terms] OR "collateral effects"
[Title/Abstract]))
|
|
Medline complete (EBSCO)
|
TX ((orthodontics OR orthodontic OR "orthodontic patients" OR "orthodontic treatment"
OR "periodontal index" OR "periodontal diseases" OR periodontitis OR periodontics
OR "periodontal patients" OR periodontal OR "periodontal treatment" OR gingivitis
OR gingiva OR "periodontitis patients" OR caries OR “carious dentin” OR “caries disease”
OR “streptococcus mutans” OR streptococcus) AND TX (photochemotherapy OR "photodynamic
therapy") AND TX ("anti-infective agents" OR antimicrobial OR bactericidal OR bacteriostatic
OR "anti-inflammatory agents" OR "antiviral agents" OR "antifungal agents" OR “cariogenic
biofilm” OR biofilms OR apoptosis OR efficacy OR "side effects" OR "adverse effects"
OR "collateral effects"))
|
|
Web of Science (Clarivate Analytics)
|
TS = ((orthodontics OR orthodontic OR "orthodontic patients" OR "orthodontic treatment"
OR "periodontal index" OR "periodontal diseases" OR periodontitis OR periodontics
OR "periodontal patients" OR periodontal OR "periodontal treatment" OR gingivitis
OR gingiva OR "periodontitis patients" OR caries OR “carious dentin” OR “caries disease”
OR “streptococcus mutans” OR streptococcus) AND (photochemotherapy OR "photodynamic
therapy") AND ("anti-infective agents" OR antimicrobial OR bactericidal OR bacteriostatic
OR "anti-inflammatory agents" OR "antiviral agents" OR "antifungal agents" OR “cariogenic
biofilm” OR biofilms OR apoptosis OR efficacy OR "side effects" OR "adverse effects"
OR "collateral effects"))
|
|
Scopus (Elsevier)
|
TITLE-ABS-KEY ((orthodontics OR orthodontic OR "orthodontic patients" OR "orthodontic
treatment" OR "periodontal index" OR "periodontal diseases" OR periodontitis OR periodontics
OR "periodontal patients" OR periodontal OR "periodontal treatment" OR gingivitis
OR gingiva OR "periodontitis patients" OR caries OR “carious dentin” OR “caries disease”
OR “streptococcus mutans” OR streptococcus) AND TITLE-ABS-KEY (photochemotherapy OR
"photodynamic therapy") AND TITLE-ABS-KEY ("anti-infective agents" OR antimicrobial
OR bactericidal OR bacteriostatic OR "anti-inflammatory agents" OR "antiviral agents"
OR "antifungal agents" OR “cariogenic biofilm” OR biofilms OR apoptosis OR efficacy
OR "side effects" OR "adverse effects" OR "collateral effects"))
|
|
Cochrane (database for systematic reviews, CENTRAL, trials, protocols)
|
#1 orthodontics OR #2 orthodontic OR #3 "orthodontic patients" OR #4 "orthodontic
treatment" OR #5 "periodontal index" OR #6 "periodontal diseases" OR #7 periodontitis
OR #8 periodontics OR #9 "periodontal patients" OR #10 periodontal OR #11 "periodontal
treatment" OR #12 gingivitis OR #13 gingiva OR #14 "periodontitis patients" OR #15
caries OR #16 “carious dentin” OR #17 “caries disease” OR #18 “streptococcus mutans”
OR #19 streptococcus AND #20 photochemotherapy OR #21 "photodynamic therapy" AND #22
"anti-infective agents" OR #23 antimicrobial OR #24 bactericidal OR #25 bacteriostatic
OR #26 "anti-inflammatory agents" OR #27 "antiviral agents" OR #28 "antifungal agents"
OR #29 “cariogenic biofilm” OR #30 biofilms OR #31 apoptosis OR #32 efficacy OR #33
"side effects" OR #34 "adverse effects" OR #35 "collateral effects"
#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR
#14 OR #15 OR #16 OR #17 OR #18 OR #19 AND #20 OR #21 AND #22 OR #23 OR #24 OR #25
OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35
|
|
Trials Central (http://www.isrctn.com)
|
Photodynamic therapy
|
|
Clinical Trials (https://clinicaltrials.gov/)
|
Orthodontics and photodynamic therapy
|
This systematic review was performed in accordance with the reference items for evaluating
articles in systematic review and meta-analysis (Preferred Reporting Items for Systematic
Review and Meta-Analysis [PRISMA]),[16]
[17] and extension statement for reports of NMAs.[18] The protocol for this NMA review has been registered on the PROSPERO platform (crd.york.ac.uk/prospero)
under ID number:CRD42024504640.
Search Strategy
An electronic search was performed in the following databases until February 3, 2025,
without limitation of year or language: PubMed (Medline), Scopus, Web of Science,
Medline Complete (EBSCO), Cochrane (Database for Systematic Review, CENTRAL and Protocols),
and gray literature through clinical trials. The search strategies are described in
[Table 1].
Article Eligibility Criteria
Two researchers (R.A.V. and R.L.S.) independently made the selections from the abstracts,
titles, and full texts, according to the eligibility criteria ([Table 1]). Discrepancies were resolved by discussion and consensus.[19]
[20]
[21] In case of disagreements between the two evaluators and a consensus could not be
reached, a third evaluator (C.M.P.) was consulted.
Quality Assessment and Risk of Bias
Two independent reviewers assessed the risk of bias of the included studies using
the Cochrane Risk of Bias for Non-Randomized Controlled Trials guidelines (ROBINS-I)
tool and the Cochrane Risk of Bias for Randomized Controlled Trials (RoB2) tool.[22]
[23]
The risk of bias for nonrandomized clinical trials (NRCTs) assessed by ROBINS-I analyzed
the following seven domains: bias due to confounding; bias in the selection of study
participants; bias in the classification of interventions; bias due to deviations
from the intended intervention; bias due to lack of data; bias in measuring results;
and bias in the selection of the reported outcome. The overall risk of bias of individual
studies was classified as low (if all domains were considered to be at low risk of
bias), moderate (if one or more domains were at moderate risk of bias), severe (if
one or more domains were at severe risk of bias), critical (if one or more domains
present a critical risk of bias).
Randomized controlled trials (RCTs) assessed by RoB2 analyzed five domains: randomization
process, deviations from intended interventions, missing outcome data, outcome measurement,
and selection of reported outcome) as “low risk,” “unclear risk,” or “high risk,”
and disagreements were again checked by a third evaluator (C.M.P.).
Extraction and Data Analysis
Two independent reviewers extracted the data. Disagreements were resolved through
discussion until a consensus was reached.[24] The primary outcome was decreased gingival inflammation, encompassing decreased
periodontal pocket probing depth (PD), dental Plaque Index (PI), Gingival Bleeding
Index (GBI), and Gingival Index (GI). The secondary outcomes were the decrease in
oral microbiological culture (OMC) and levels of immunoinflammatory cytokines in the
gingival crevicular fluid (GCF).
Grades of Recommendations Assessment, Development, and Evaluation
Owing to the low heterogeneity of the data and methodology of the included studies,
it was possible to gather information necessary for a meta-analysis. For heterogeneous
data, a narrative synthesis was approached, using a table summarizing the results
according to the Grades of Recommendations Assessment, Development and Evaluation
(GRADE) pro system. GRADE used the Cochrane Risk of Bias for Randomized Controlled
Trials (RoB2) tool to assess the certainty of evidence for narrative synthesis.[17]
[25] The use of these tools was important to allow a better comparison of the evidence
and the degree of certainty for recommending decision-making.[17]
[25]
The GRADE approach assessed the number of included studies, study designs, risk of
bias, inconsistency, indirect evidence, imprecision, and publication bias. Depending
on the severity of the limitation in each of these categories, the evidence was downgraded
by 1 or 2 levels. Based on this assessment, the certainty of the outcome assessment
could be very low, low, moderate, or high quality. In NRCTs, outcomes that demonstrated
a large magnitude effect were upgraded by 1 or 2 levels. For CCT, the initial certainty
level was high.[17]
[23]
Data Synthesis and Statistical Analysis
For each continuous outcome, we collected the mean and standard deviation (SD) at
baseline and each time point. We calculated the mean difference (MD) for each intervention
(baseline from the last time point). The SD was calculated from the standard error
(SE) obtained in the Review Manager calculator (desktop, version 5.4) for each intervention.
Based on the SE, we obtained the SD for each intervention considering the following
formula: 
.[26]
There were four treatments: US, PDT + US, PMB + US, and PDT, the latter being the
reference intervention.
First, data were entered in the MetaInsight online software version 5.1.0 to run a
Bayesian[27] NMA for analysis using Markov Chain Monte Carlo simulation for the following outcomes:
PD, PI, GBI, Porphyromonas gingivalis, and Fusobacterium nucleatum. We considered the MD change from baseline and SD for each intervention to calculate
the final effect estimate (EE): MD and respective credible intervals (95% credible
intervals [CrI]) for comparing two interventions, as the studies used the same scale
to measure the outcomes. We obtained each network's deviance information criterion
(DIC) using fixed and random models.[27]
[28] The final model was chosen based on the lowest DIC ([Supplemental Material] [available in the online version only]). For PD, PI, and P. gingivalis, we used four chains with 1:20.000 interactions. For GBI and F. nucleatum, there were four chains with 5.001:25.000 interactions.
Sequentially, we obtained the geometry for each outcome. Then, we assessed the convergence
model based on trace plots and time-series plots. Incoherence was tested by comparing
direct estimates with the indirect estimates for each comparison through the node-splitting
technique.[28] The node-splitting technique did not run for PD and F. nucleatum due to the lack of direct and indirect evidence in a closed loop. In these cases,
we checked for inconsistency throughout the global inconsistency plot. League tables
were obtained for each outcome. Finally, the ranking was calculated using the surface
under the cumulative ranking (SUCRA). As the ranking probability can lead to misleading
conclusions, we followed the certainty of the evidence for the interpretation of data.[29]
As all studies had a high or unclear risk of bias, we did not run a sensitive analysis
for risk of bias. Risk of bias was considered in assessing the certainty of the evidence
(please see the item below). Instead, the robustness of findings was checked by recalculating
each NMA using only the last time point of each intervention to avoid inputting the
SD for all interventions. Both NMAs rendered similar results. Finally, we will present
here the results of the NMAs calculated using the last time point only. The network
results considering the MD for the intervention are presented in the [Supplementary Material] (available in the online version only), as well as the results for change from baseline
([Supplemental Material] [available in the online version only]).
A pairwise random effect model meta-analysis was run for the outcomes that did not
have a connected network: Gingival Index and Tannerella forsythia. We used the final time point and SD to calculate MD and 95% CI between interventions.
Finally, three outcomes did not have at least two studies comparing the same interventions:
P. intermedia, IL-6, and TNF-α. Therefore, we plotted the studies' SE in a series of forest plots
using a fixed model. As the authors used the same scale for all outcomes, we considered
MD as the final SE, except for IL-6 and TNF-α. The means and SD of these two outcomes
varied on a scale from 10 to 100. In these cases, we calculated the final SE as a
standardized mean difference (SMD) and 95% CI to make the SE of forest plots comparable.
The Review Manager version 5.4 was used for the pairwise comparisons.
Results
Study Selection
After screening titles and abstracts of 1,528 articles, 65 potentially eligible articles
were selected for full-text analysis, of which 6 randomized clinical trials (RCTs)[12]
[30]
[31]
[32]
[33]
[34] were included ([Fig. 1]).
Fig. 1 Flowchart of study selection showing the synthesis of the systematic review, in accordance
with PRISMA guidelines.
Characteristics of the Studies
The studies evaluated a total of 173 patients with an average age ranging between
12 and 19 years ([Table 2]). The studies were performed between 2015 and 2020 in Spain[12]
[33] and Saudi Arabia.[30]
[31]
[32]
[34]
Table 2
Main characteristics of the included studies
|
Authors (year)
Country
|
Study design
Follow-up
|
No. of treated patients (P)
per group
|
Age range (mean)
Gender eligibility criteria
|
Treatment procedure
|
Conclusions
|
Mean difference in PD between baseline and final follow-up (mm)/p-value
|
Mean difference in FMPI between baseline and final follow-up (%)/p-value
|
Mean difference in FMBI between baseline and final follow-up (%)/p-value
|
Mean difference in GI between baseline and final follow-up (%)/p-value
|
Mean difference in OMC between baseline and final follow-up, (CFU/mL)/p-value
|
Mean difference in GCF between baseline and final follow-up, (pg/mL)/p-value
|
|
Gomez et al (2018)[33]
[Spain]
|
RCT
(full-mouth)
270 d
|
10P (T)
10P (C)
|
12–18
♂7/♀3 (T)
15 ± 1.8y
♂7/♀3 (C)
14.2 ± 1.3y
OT ≥15 m
|
Day 0; 15; 30; 45; 90; 180; 270
PDT 670 nm; 60 s/tooth; methylene blue
0.005% (T)
US (Sonic Flex, Scaler no. 5);
full-mouth (C)
|
Both US and PDT improved in a similar way clinical outcomes and microbiological counts
during the orthodontic treatment in adolescents with fixed devices. A progression
of gingival inflammation or enamel demineralization is not to be expected within 9
and 6 months, respectively, after repeated implementation of prophylactic procedures
when there is sufficient oral hygiene practice
|
0.90 ± 0.00 (T)
p > 0.05
0.40 ± 0.1 (C)
p > 0.05
|
24 ± 4 (T)
p > 0.05
35 ± 14 (C)
p > 0.05
|
9.8 ± 2.8 (T)
p > 0.05
6.9 ± 2.3 (C)
p > 0.05
|
NR
|
P. gingivalis
1.63 ± 0.27 (T)
p > 0.05
0.67 ± 1.57 (C)
p > 0.05
P. intermedia
1.30 ± 1.09 (T)
p > 0.05
0.99 ± 0.0 (C)
p > 0.05
S. mutans
0.09 ± 0.06 (T)
p > 0.05
0.22 ± 0.04 (C)
p > 0.05
|
NR
|
|
Abellan et al (2019)[12] [Spain]
|
RCT
(full-mouth)
270 d
|
10P (T)
10P (C)
|
12–18
(14.6 ± 1.6)
♂14/♀6
OT ≥12 m
|
Day 0; 15; 30; 45; 90; 180; 270
PDT 670 nm; 60 s/tooth; methylene blue
0.005% (T)
US (Sonic Flex, Scaler no. 5);
full-mouth (C)
|
Both PDT and ultrasonic scaling are safe and effective treatment methods for gingival
inflammation induced by fixed orthodontic appliances. In terms of clinical, microbiological,
and anti-inflammatory outcomes, PDT was slightly more effective than the implementation
of the US, allowing the extension of the benefits for a longer period
|
0.70 ± 0.00 (T)
p > 0.05
0.40 ± 0.15 (C)
p > 0.05
|
[a]0.59 ± 0.01 (T)
p > 0.05
[a]0.59 ± 0.08 (C)
p > 0.05
[a]Scale: Silness and Loe,[23] (scores: 0–3)
|
NR
|
0.90 ± 0.28(T)
p > 0.05
0.53 ± 0.05 (C)
p > 0.05
|
P. gingivalis
1.63 ± 0.27 (T)
p > 0.05
0.67 ± 1.57 (C)
p > 0.05
P. intermedia
1.30 ± 1.09 (T)
p > 0.05
0.99 ± 0.0 (C)
p > 0.05
F. nucleatum
0.45 ± 0.75 (T)
p > 0.05
0.03 ± 0.10 (C)
p > 0.05
|
IL-1β
25.79 ± 33.95 (T)
p > 0.05
19.05 ± 13.07 (C)
p > 0.05
IL-10
5.94 ± 3.28 (T)
p < 0.05
4.71 ± 1.76 (C)
p < 0.05
TNF-α
2.27 ± 0.94 (T)
p > 0.05
2.20 ± 0.89 (C)
p > 0.05
|
|
Alqerban (2020)[31] [Saudi Arábia]
|
RCT
(full-mouth)
60 d
|
15P (T)
15P (C1)
15P (C2)
|
12–19
♂4/♀11 (T)
14.7 ± 0.8y
♂6/♀9 (C1)
16.2 ± 0.9y
♂3/♀12 (C2)
15.8 ± 0.7y
OT ≥8 m
|
Day 0; 30; 60
PDT (670 nm, 60 s/tooth; methylene blue
0.005%) + US (T)
PBM (diode laser 810 nm) + US (C1)
US + manual curette; full-mouth (C2)
|
PDT and PBM showed similar improvement in gingival inflammatory and microbiological
parameters compared to US. PDT assisted in a modest reduction of hBD-2 in patients
undergoing fixed orthodontic treatment
|
0.78 ± 0.00 (T)
p > 0.05
0.31 ± 0.10 (C1)
p > 0.05
0.30 ± 0.0 (C2)
p > 0.05
|
39 ± 1 (T)
p < 0.05
39 ± 0 (C1)
p > 0.05
31 ± 0 (C2)
p < 0.05
|
12 ± 0 (T)
p < 0.05
14 ± 0 (C1)
p < 0.05
15 ± 2 (C2)
p < 0.05
|
NR
|
T. denticola
0.60 ± 0.60 (T)
p < 0.05
0.42 ± 0.20 (C1)
p > 0.05
0.37 ± 0.10 (C2)
p > 0.05
F. nucleatum
0.23 ± 0.20 (T)
p < 0.05
0.26 ± 0.10 (C1)
p < 0.05
0.10 ± 0.30 (C2)
p > 0.05
|
hBD-2
25 ± 1 (T)
p < 0.05
17 ± 2 (C1)
p > 0.05
6 ± 1 (C2)
p > 0.05
|
|
Baeshen et al (2020)[32] [Saudi Arábia]
|
RCT
(full-mouth)
28 d
|
15P (T)
15P (C)
|
14-19
♂5/♀10(T)
16.1 ± 1.4y
♂6/♀9(C)
15.9 ± 1.3y
OT ≥8 m
|
Day 0; 7; 28
PDT (670 nm, 120 s/tooth; methylene blue
0,005%)
+ US (T)
US (Cavitron, Scaler); full-mouth (C)
|
PDT has a positive effect in significantly reducing the periodontal microbial load
in established gingivitis in adolescent patients undergoing fixed orthodontic treatment.
|
0.60 ± 0.10(T)
p > 0.05
0.70 ± 0.10 (C)
p > 0.05
|
25.5 ± 11.9 (T)
p < 0.05
25.3 ± 11.1 (C)
p < 0.05
|
42.6 ± 16.5 (T)
p < 0.05
30.3 ± 12.6 (C)
p < 0.05
|
NR
|
P. gingivalis
0.98 ± 0.58 (T)
p < 0.05
0.60 ± 0.33 (C)
p < 0.05
F. forsythia
0.83 ± 0.19 (T)
p < 0.05
0.41 ± 0.32 (C)
p < 0.05
|
TNF-α
19.9 ± 51.8 (T)
p < 0.05
51 ± 25.2 (C)
p < 0.05
IL-6
35.4 ± 18.0 (T)
p < 0.05
34.8 ± 4.9 (C)
p < 0.05
|
|
Malik and Alkadhi (2020)[34] [Saudi Arábia]
|
RCT
(full-mouth)
180 d
|
18P (T)
18P (C)
|
16–17
♂10/♀8(T)
16.6 ± 0.5y
♂9/♀9(C)
16.8 ± 0.4y
OT ≥9 m
|
Day 0; 180
PDT (660 nm, 60 s/tooth;
methylene blue
0,005%) (T)
US (W Dental, Scaler); full-mouth (C)
|
The PDT is a useful adjuvant to US in reducing whole salivary oral yeast counts among
adolescents undergoing orthodontic treatment. In the short-term, US with and without
PDT is useful in reducing GI in adolescents undergoing orthodontic treatment
|
NR
|
NR
|
NR
|
1.8 ± 0.22 (T)
p < 0.05
1.6 ± 0.05 (C)
p < 0.05
|
Total
81.8 ± 5.9 (T)
p < 0.05
24.2 ± 0.0 (C)
p < 0.05
|
NR
|
|
Al Nazeh et al (2020)[30]
[Saudi Arabia]
|
RCT (full-mouth)
28 d
|
11P (T)
11P (C)
|
16-18
(17.5y)
♂4/♀7(T)
♂5/♀6(C)
OT ≥9 m
|
Day 0; 7; 28
PDT (670 nm, 60 s/tooth;
methylene blue
0.005%) + US (T)
US (Scaler); full-mouth (C)
|
PDT was effective in significantly reducing periodontal pathogens in established gingivitis
lesions in adolescent patients undergoing fixed orthodontic treatment in the short
term
|
NR
|
41.1 ± 19.8 (C)
p < 0.05
35.4 ± 14.8 (T)
p < 0.05
|
39.8 ± 23.5 (C)
p < 0.05
48 ± 9.6 (T)
p < 0.05
|
NR
|
P. gingivalis
1.19 ± 0.62 (T)
p < 0.05
0.26 ± 0.50 (C)
p > 0.05
F. forsythia
1.11 ± 0.10 (T)
p < 0.05
0.48 ± 0.35 (C)
p > 0.05
|
NR
|
Abbreviations: C, control; FMBS, full-mouth bleeding index (%, scale, dichotomous);
FMPI, full-mouth Plaque Index (%, scale, dichotomous); GCF, gingival crevicular fluid;
GI, Gingival Index (%, scale, dichotomous); NR, not reported; OMC, oral microbiological
culture; OT, orthodontic treatment; PBM, photobiomodulation; PD, probing depth; PDT,
photodynamic therapy; RCT, randomized clinical trial; T, treatment; US, ultrasonic
scaling.
a Scale: Silness and Loe 23, (scores: 0-3).
All the studies[12]
[30]
[31]
[32]
[33]
[34] evaluated the decrease in periodontopathogenic flora. Five RCTs[12]
[30]
[31]
[32]
[33] evaluated bag PD and full mouth PI.
Five RCTs[12]
[30]
[31]
[32]
[33] evaluated the full-mouth GBI. Three RCTs[12]
[31]
[34] evaluated the Gingival Index (GI).
Inflammatory biomarkers were evaluated in two RCTs in GCF.[12]
[32] In the evaluation of GCF, two RCTs[12]
[32] evaluated cytokines including IL-1, IL-6, IL-10, and TNF.[12] PDT in five RCTs[12]
[31]
[32]
[33]
[34] was associated with methylene blue.
As a control, three RCTs used US,[12]
[30]
[33] one RCT used US and laser,[31] and two RCTs used manual scraping.[32]
[34]
All the studies[12]
[30]
[31]
[32]
[33]
[34] used the full-mouth system for treatment. And the duration of treatments varies
by 4 weeks,[30]
[31]
[32] 8 weeks,[31] 6 months,[34] and up to 9 months.[33]
Bias Risk
Two studies[12]
[33] presented an unclear risk of bias, and four studies[30]
[31]
[32]
[34] were identified as high risk of bias ([Fig. 2A]). In the studies[12]
[33] with an unclear risk of bias, the concern observed was the lack of information on
deviations from the intended interventions.
Fig. 2 (A) Risk of bias summary of included randomized trials with the Cochrane Risk of Bias
Assessment Tool version 2 (RoB2). (B) Summary of findings (SoF) table according to the Grades of Recommendations, Assessment,
Development, and Evaluation (GRADE) approach for narrative synthesis.
Studies with high risk of bias[30]
[31]
[32]
[34] occurred mainly due to the lack of information on deviations from the intended interventions,
such as blinding of patients and operators. Risk of bias in outcome measurement, such
as lack of assessor blinding, was another concern in studies at high risk of bias.
Results of the Included Study
Studies Not Eligible for Meta-Analysis
A summary in [Fig. 2B] describes the results and certainty of evidence using the GRADE approach for narrative
synthesis of studies not included in either the paired meta-analysis or the NMA for
reducing gingival inflammation on PI, OMC, and GCF outcomes.
Four RCTs[12]
[31]
[33]
[34] were assessed. One RCT[12] evaluated the effectiveness of PDT on PI control with moderate evidence certainty.
Three RCTs[31]
[33]
[34] evaluated the effectiveness of PDT on microbiological culture control, but the evidence
was very uncertain. Two RCTs[12]
[31] evaluated the efficacy of PDT on the control of IL-1b, IL-1ra, PGF-2, IL-10, and
hBD-2, but the certainty of the evidence was low.
Certainty of the Evidence of the Narrative Synthesis
In the GRADE approach to narrative synthesis, the certainty of evidence was very low
to moderate due to issues of risk of bias and imprecision ([Fig. 2B]).
Synthesis and Certainty of Evidence from Studies in Forest Graphics
Three outcomes did not have at least two studies comparing the same interventions:
Prevotella intermedia, IL-6, and TNF-α. Therefore, we plotted the final effect estimate (EE) of the studies
on a series of forest plots using a fixed model.
[Fig. 3 (A, B)] shows the single study SMD for IL-6. Efficacy was similar for the comparison between
PDT and US (SMD, 0.05; 95% CI, −0.83 to 0.92; very low certainty; [Fig. 3A], [Table 3]); and between PDT + US and US (SMD, 0.05; 95% CI, −0.66 to 0.77; very low certainty)
for decreased IL-6 ([Fig. 3B], [Table 3]).
Fig. 3 (A) Forest plot showing the standardized mean difference (SMD) for IL-6 decrease between
PDT therapy and US therapy. (B) Forest plot showing the standardized mean difference (SMD) for IL-6 decrease between
PDT + US therapy and US therapy. (C) Forest plot showing the standardized mean difference (SMD) for decreased TNF-α between
PDT therapy and US therapy. (D) Forest plot showing the standardized mean difference (SMD) for decreased TNF-α between
PDT + US therapy and US therapy. (E) Forest plot showing the mean difference (MD) for P. intermedia decrease between PDT therapy and US therapy. (F) Meta-analysis comparing photodynamic therapy (PDT) with ultrasound (US) therapy
to reduce the Gingival Index. (G) Meta-analysis comparing PDT + US therapy with US therapy to reduce T. forythia. (H) Network geometries. (A–C) Primary outcome, Probing Depth, Plaque Index, Gingival Bleeding Index; (D–E) Secondary outcome, P. gingivalis, F. nucleatum. (A) Probing Depth treatment network, (B) Plaque Index Treatment Network, (C) Gingival Bleeding Index treatment network, (D) treatment network for P. gingivalis, (E) treatment network for F. nucleatum. PDT, photodynamic therapy; PMB, photobiomodulation; US, ultrasonic scaling.
Table 3
Average difference (95% CI) and certainty of evidence of treatments for primary (Probing
Depth, Plaque Index, Gingival Beeding Index) and secondary (P. gingivalis, F. nucleatum) outcomes
|
Probing Depth (4 trials)[a]
|
NMA
|
|
MD (95% CI)[b]
|
Evidence certainty
|
|
Treatment comparison
|
|
|
|
A
|
|
B
|
|
|
|
PDT
|
Vs.
|
PDT-US
|
−0.45 (−0.74, −0.17)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
PMB-US
|
−0.06 (−0.31, 0.19)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
US
|
0.15 (−0.09, 0.38)
|
Very low[d]
[e]
|
|
PDT-US
|
Vs.
|
PMB-US
|
0.39 (0.23, 0.55)
|
Very low[c]
[d]
|
|
PDT-US
|
Vs.
|
US
|
0.6 (0.44, 0.76)
|
Very low[c]
[d]
|
|
PMB-US
|
Vs.
|
US
|
0.21 (0.13, 0.29)
|
Very low[c]
[d]
|
|
Plaque Index (5 trials)
[f]
|
|
|
|
PDT
|
Vs.
|
PDT-US
|
−2.78 (−9.02, 3.51)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
PMB-US
|
−0.8 (−10.45, 8.73)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
US
|
−4.24 (−12.11, 3.69)
|
Very low[d]
[e]
|
|
PDT-US
|
Vs.
|
PMB-US
|
1.94 (−5.65, 9.49)
|
Very low[c]
[d]
|
|
PDT-US
|
Vs.
|
US
|
−1.5 (−6.96, 4.04)
|
Very low[c]
[d]
|
|
PMB-US
|
Vs.
|
US
|
−3.45 (−10.6, 3.77)
|
Very low[c]
[d]
|
|
Gingival Bleeding Index (4 trials)
[g]
|
|
|
|
PDT
|
Vs.
|
PDT-US
|
1.39 (−7.09, 10.32)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
PMB-US
|
3.35 (−7.05, 15.02)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
US
|
1.68 (−5.29, 10.18)
|
Very low[d]
[e]
|
|
PDT-US
|
Vs.
|
PMB-US
|
1.99 (−6.15, 10.81)
|
Very low[c]
[d]
|
|
PDT-US
|
Vs.
|
US
|
0.42 (−5.67, 7.18)
|
Very low[c]
[d]
|
|
PMB-US
|
Vs.
|
US
|
−1.65 (−10.17, 7.08)
|
Very low[c]
[d]
|
|
Porphyromonas gingivalis
(4 trials)
[h]
|
|
|
PDT
|
Vs.
|
PDT-US
|
−0.47 (−1.47, 0.53)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
US
|
0.03 (−1.17, 1.23)
|
Very low[d]
[e]
|
|
PDT-US
|
Vs.
|
US
|
0.5 (−0.44, 1.44)
|
Very low[c]
[d]
|
|
Fusobacterium nucleatum
(2 trials)
[i]
|
|
|
PDT
|
Vs.
|
PDT-US
|
0.57 (−0.22, 1.36)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
PMB-US
|
0.62 (−0.18, 1.41)
|
Very low[c]
[d]
|
|
PDT
|
Vs.
|
US
|
0.65 (0, 1.29)
|
Very low[d]
[e]
|
|
PDT-US
|
Vs.
|
PMB-US
|
0.05 (−0.38, 0.48)
|
Very low[c]
[d]
|
|
PDT-US
|
Vs.
|
US
|
0.08 (−0.38, 0.54)
|
Very low[c]
[d]
|
|
PMB-US
|
Vs.
|
US
|
0.03 (−0.43, 0.49)
|
Very low[c]
[d]
|
Abbreviations: CI, confidence interval; DM, average difference; NMA, network meta-analysis;
PDT, photodynamic therapy; PDT-US, photodynamic therapy—ultrasonic scaling; PMB-US,
photobiomodulation—ultrasonic scaling; US, ultrasonic scaling.
Notes: Data are presented per NMA. Fixed effect model was used, except Gingival Bleeding
Index, which used a random effect model.
None of the estimates was demoted by intransitivity.
a Four trials: Gómez et al[33]; Abellán et al[12]; Alqerban[31]; Baeshen et al.[32]
b Positive values favor treatment A, and negative values favor treatment B.
c Certainty in evidence downgraded by 2 levels due to very serious risk of bias.
d Certainty in evidence downgraded by 1 level due to serious risk of bias.
e Certainty in evidence downgraded by 2 levels due to very serious imprecision.
f Five trials: Gómez et al[33]; Abellán et al[12]; Alqerban[31]; Baeshen et al[32]; Al Nazeh et al.[30]
g Four trials: Gómez et al[33]; Alqerban[31]; Baeshen et al[32]; Al Nazeh et al.[30]
h Four trials: Gómez et al[33]; Abellán et al[12]; Baeshen et al[32]; Al Nazeh et al.[30]
i Two trials: Abellán et al[12]; Alqerban.[31]
[Fig. 3 (C, D)] shows the SMD of the single study, demonstrating that orthodontic patients undergoing
a comparison between PDT and US (SMD, −0.29; 95% CI, −1.17 to 0.60; very low certainty;
[Fig. 3C], [Table 3]); and between PDT + US and US (SMD, 0.38; 95% CI, −0.34 to 1.11; very low certainty)
for decreasing TNF-α demonstrated uncertainty in the efficacy of PDT ([Fig. 3D], [Table 3]).
[Fig. 3E] shows the single-study MD for P. intermedia reduction. PDT treatment had superior or similar efficacy to US (MD, −0.36; 95% CI,
−2.05 to 1.33; with very low certainty; [Fig. 3E], [Table 3]).
Meta-Analysis Results
Paired Meta-Analysis
A paired random effect model meta-analysis was performed for the outcomes that did
not have a connected network: GI and Tannerella forsythia. We used the final moment and the SD to calculate MD and 95% CI between interventions.
[Fig. 3 (F and G)] shows the MD of paired studies for the GI. Treatment with PDT (MD, −0.10; 95% CI,
−0.14 to −0.05) had a greater chance of being effective in reducing the GI compared
with treatment with US (very low certainty; [Fig. 3F], [Table 3]).
PDT + US was superior or similar compared with US (MD, −0.36; 95% CI, −0.87 to 0.14;
very low certainty) for decreasing T. forythia ([Fig. 3G], [Table 3]).
Network Meta-Analysis
[Table 3] and [Fig. 3H] present the “summary of findings,”, the network geometries with the four groups
of treatments for the primary outcomes (PD, PI, GBI), and secondary outcomes (P. gingivalis, F. nucleatum). All treatments had very low certainty, which shows a lack of certainty regarding
their effectiveness. Furthermore, the 95% CrI crosses the line of null effect for
the treatments in the outcomes of PI, GBI, and P. gingivalis. For PD, this occurs with the US-PDT treatment (0.15: −0.09, 0.38). In the outcome
of F. nucleatum, all treatments show this pattern except for the US-PDT (0.65: 0.0, 1.29), indicating
very serious inaccuracy.
Discussion
Eligible studies that demonstrated different methodologies[12] or single-study outcomes[12]
[31]
[33]
[34] were evaluated using the GRADE approach.[25] One study[12] used the numerical scale (score: 0–3) of Silness and Loe[35] for the PI, unlike the other studies[30]
[31]
[32]
[33]
[34] that employed the dichotomous scale. PDT demonstrated similar efficacy to US therapy
with moderate certainty due to imprecision. In the outcome of microbiological culture,[31]
[33]
[34] for Streptococcus mutans,[33] PDT demonstrated similar efficacy to US; for Treponema denticola
[31] and the total count of microorganisms,[34] PDT demonstrated superior efficacy to US[31]
[33]
[34]; these findings do not corroborate a previous review[14] that reported the absence of significant differences between the intervention and
comparison groups for T. denticola. However, the certainty of the evidence was very low[31]
[33]
[34] due to the risk of bias and imprecision (size of the sample).
Studies[12]
[31] evaluating cytokines (IL-1b, IL-1ra, PGF-2, IL-10, hBD-2) in GCF presented individual
assessments without comparative analysis. PDT[12] may demonstrate similar efficacy to US therapy in controlling inflammatory mediators
(IL-1b, IL-1ra). Fibroblast growth factor (PGF-2) may show superior efficacy compared
to IL-10,[12] and superior to US and/or 810-nm diode laser in regulating antimicrobial peptides
such as human β-defensins (hBD-2).[31] However, these findings carry low certainty due to the risk of bias (methodological
problems) and imprecision related to the sample size.
Analysis of IL-6 demonstrated that PDT treatments,[8] US,[12]
[28] and PDT + US[32] had similar efficacy, that is, similar effects in reducing this cytokine. To reduce
TNF-α, PDT[8] treatments, US,[12]
[28] and PDT + US[32] demonstrated uncertainty in treatments. P. intermedia demonstrated uncertainty in treatments. For P. intermedia,[33] the treatments using PDT and US had similar effectiveness, which corroborates the
review findings of one study,[14] which reported no significant differences between the intervention and comparison
groups.
In a previous study,[36] the authors reported that levels of microbial pathogens such as P. intermedia
[33] as well as the inflammatory cytokines IL-6 and TNF-α[12]
[32] demonstrated a significant decrease in inflammatory markers after treatment with
PDT compared with other treatments. Another study[14] with meta-analysis found no significant differences between treatment groups for
cytokine outcomes in GCF and for bacterial analysis. These findings were not fully
consistent with the present study in presenting the effectiveness of PDT treatment
and the certainty of the evidence related to these outcomes and other treatments.
Bahrami et al[36] reported limitations that influence outcomes, such as variations in protocols and
techniques,[12]
[32] which align with our findings. Previous studies[12]
[32] evaluating the cytokines IL-6 and TNF-α reported PDT techniques applied for 60 seconds/tooth[12] and 120 seconds/tooth.[32] Follow-up assessments were conducted at various intervals, including 0, 7, and 28
days[32] as well as 0, 15, 30, 45, 90, 180, and 270 days.[12] In this study, the meta-analysis demonstrated that PDT was more effective than the
US therapy in reducing the GI,[12]
[34] but this evidence is uncertain. However, PDT + US and US therapy alone showed similar
effectiveness in reducing T. forythia.[30]
[32] A recent study[14] did not find significant differences between the intervention and comparison groups
for these outcomes. Although the present study found a higher likelihood of efficacy
for PDT in improving the GI, the certainty of the evidence remains very low.[12]
[30]
[32]
[34]
NMA results demonstrated that all treatments were similar for reducing PI, GBI, PD,
P. gingivalis, and F. nucleatum. When considering the effectiveness of PDT treatment compared with other treatments,
the 95% CI was wide and crossed the null effect line for the outcomes of PI, GBI,
P. gingivalis, and PD. F. nucleatum presented imprecise results. The comparison between PDT and US treatment (0.65: 0.0,
1.29) favored PDT, but with very low evidence certainty, due to very serious imprecision.
The findings of the present study corroborate those of Shafaee et al[14] who reported not having observed a significant difference between PDT and US treatments,
or low-intensity laser therapy for reducing PI, regardless of the evaluation time.
In a previous study,[14] PDT was significantly more effective than US in the reduction of PD; however, this
difference seemed to be clinically insignificant. NMA's present study showed that
the data showed uncertainty in treatments, with chances of efficacy for PDT and also
for US. For GBI and P. gingivalis, the findings of this study corroborate with study,[14] which found no significant difference for PDT compared with other treatments.
Overall, the results of this NMA showed some divergences from a review[36] and a previous MA,[14] which we credit to the different analyses performed, and to the rigor of the inclusion
and exclusion criteria of the eligible studies evaluated. The previous reviews[14]
[36] focused on evaluating white spot lesions and gingivitis. A study without meta-analysis,[36] which restricted the criteria to only the English language and patients over 15
years old, did not report on smoking patients, pregnant women, erupting teeth, orthodontic
retreatment, use of antibiotics, and anti-inflammatories. In another study with MA,[14] the authors were also not clear enough with the inclusion and exclusion criteria,
specifying only the inclusion of low-intensity laser, PDT: scrapers, rinses, prophylactic
treatments, and varnishes. We understand that the inclusion and exclusion criteria
are important items to reduce the confounding and heterogeneity factors listed by
these studies.[14]
[36] The MA study[14] reported that the designs of the studies by Al Nazeh et al[30] and Alqerban[31] were split-mouth, but considering the methodology of the studies and their distribution
of groups, we showed that they were full-mouth assessment groups per treatment, which
could impact the results in some way.
The therapeutic capacity of PDT is attributed to photodamage caused by reactive oxygen
species,[14]
[32]
[36] which are cytotoxic substances that can damage the bacterial cell membrane and DNA,
leading to cell death.[37]
[38] Studies have reported that PDT is a safe therapy.[13]
[39] The potential long-term oral mucosal toxicity or phototoxicity of photosensitizers
on treatment is low. PDT allows frequent application, as it is a noninvasive procedure
without causing cumulative toxicity.[13]
[39] Photocytotoxic reactions occur only in pathological tissues, in the area of distribution
of the photosensitizer, allowing for narrowly selective destruction.[13]
[39]
[40] Methylene blue was the photosensitizer used in the eligible studies of this review;
it has a specific cationic load that allows it to connect to gram-negative bacterial
membranes, giving it high specificity to kill microorganisms.[34]
[40] It is not possible to say whether other photosensitizers would have more effective
results in orthodontic patients.
The results of this NMA demonstrated that the patient can benefit from both US and
PDT, associated or not with other treatments to control gingival inflammation, however,
with low to very low certainty for the majority of the outcomes listed, which means
that the estimate could change in future studies, and does not corroborate Bahrami
et al[36] who reported that all investigated PDT protocols were effective. However, it is
important to highlight that PDT has a high potential value in specific scenarios,
such as in patients with pacemakers or other electronic cardiac devices, who should
not receive US because the vibration can alter the functioning of these devices. Furthermore,
patients with high tooth sensitivity, active infections, ulcerated mucosal lesions,
and chlorhexidine intolerance may benefit from PDT.
The SUCRA analysis conducted in this study indicated a probability of better treatment
outcomes with PDT in reducing F. nucleatum and GBI, with combined PDT and US in reducing PD and P. gingivalis, and with US alone in reducing PI. In the coherence analysis, the studies demonstrated
global coherence ([Supplemental Material] [available in the online version only]), and coherence between direct and indirect
estimates through the knot splitting test.[41] However, with the probability presented by other therapies, it was not possible
to certify the best treatment ([Supplemental Material] [available in the online version only]). Moreover, the SUCRA ranking should not
be considered as the final decision to make an interpretation. As SUCRA does not consider
clinical thresholds and the certainty of the evidence, it can lead to misleading conclusions.[29] Instead, our study followed the certainty of the evidence to interpret data.
Regarding the comparison groups, the included studies used similar treatment protocols
for PDT (i.e., 60 seconds/tooth). However, the study by Baeshen et al[32] used 120 seconds/tooth, ranging from a single irradiation session[34] to six sessions[12]
[33] distributed over a 9-month follow-up period. Reviews identified a distinct protocol,[14]
[36] highlighting considerable variability among the studies evaluated, such as application
durations of 60 seconds,[14]
[36] 180 seconds,[14]
[36] 294 seconds,[14] and up to 10 minutes.[14]
[36]
All the studies[12]
[30]
[31]
[32]
[33]
[34] of this NMA used PDT with the diode laser (light emitting diode [LED]) as a light
source with a wavelength between 660 and 670 nm, different from other studies that
jointly evaluated wavelengths of 450 nm[14]
[36] and 640 nm,[14] which may imply heterogeneity between studies and intransitivity in data from previous
reviews.
All the studies[12]
[30]
[31]
[32]
[33]
[34] reported age symmetry. The age of the participants ranged from a minimum of 12 years
old to a maximum of 19 years old, and they were not demoted due to intransitivity.
Data from previous studies also mostly covered this age group.[14]
[36] Although this age group included only adolescents and young adults, we believe it
is not possible to clearly state whether these results are applicable to adults in
general, a concern not reported by other authors.[14]
[36]
Strengths and Limitations
The strength of this NMA lies in our rigorous methodology and well-defined inclusion/exclusion
criteria to reduce the impact of heterogeneity. This is the first NMA report on PDT
in orthodontic patients and its outcomes that allowed multiple treatments to be compared.
In addition, the quality of evidence was evaluated by the Grade of Recommendations,
Assessment, Development, and Evaluation, and all studies included were randomized.
It was impossible to perform an MA comparing the treatments for some outcomes, but
evaluated on the grid, due to the limited number of studies that exploit these treatments.
Among the limitations, we were unable to do meta-regression for time of accompaniment
due to the limited number of studies (chapter 10 of the Cochrane book)[42]—five studies were included in the NMA, and two in the pairwise meta-analysis. Another
limitation was not running a subgroup analysis in NMA, which could decrease the number
of studies per group, decreasing the sample size per group, and increasing the imprecision.
Furthermore, the forest plots (pairwise meta-analysis) do not show statistically significant
heterogeneity ([Fig. 3F]: I
2: 0%, p = 0.033; [Fig. 3G]: I2
: 0%, p = 0.73).
This review included a comprehensive search, incorporating five databases, two clinical
trials registries, and strategies. The studies were not industry-sponsored, and the
funnel plot analysis was not possible due to the small number of studies in each meta-analysis
and NMA. Studies had a small sample size, and although it can be considered a source
of publication bias, we did not rate down due to publication bias to avoid being penalized
twice for the same reason. Instead, the certainty of the evidence was rated down due
to imprecision (optimal information size not met or wide 95% CI).
We did not do a sensitivity test excluding studies at high risk of bias, due to insufficient
studies, but we circumvented this problem by lowering certainty due to the risk of
bias.
Implications for Practice and Future Research
New randomized clinical trials are expected to focus on more detailed information
on the implementation of randomization, allocation concealment, blinding, inclusion
of larger samples, and use of validated and sufficiently clear assessment tools for
different outcomes. Search for new treatment therapies or combinations of them, e.g.,
home brushing and toothpaste control.
Conclusion
The hypothesis that PDT treatment is more effective in controlling gingival inflammation
has not been confirmed. The null hypothesis was partially confirmed.
With very low to low certainty, there is no strong evidence to support PDT as a more
effective treatment for controlling gingival inflammation in orthodontic patients.
Therefore, based on the present results, the patient may benefit from either US or
PDT alone or combined with other therapies. From this perspective, the most economical
and simple US seems to be satisfactory.
