A systematic review and network meta-analysis on the optimal wavelength of low-level light therapy (LLLT) in treating knee osteoarthritis symptoms

Abstract

Objectives
To compare the efficacy of the various wavelengths of low-level light therapy (LLLT) in alleviating knee pain, dysfunction, and stiffness in patients with knee osteoarthritis (KOA), and to compare the effectiveness of LLLT versus sham treatment in reducing knee pain, dysfunction, and stiffness.

Methods
PubMed, Web of Science, EMBASE, and Cochrane Library were searched from inception to 12 December 2023. Randomized controlled trials that assessed the effects of different wavelengths of LLLT on alleviating pain of patients with KOA were included. A conventional meta-analysis and network meta-analysis were preformed, and standardized mean differences (SMD) with 95% confidence interval (CI) were calculated.

Results
Thirteen studies involving 673 participants with KOA met inclusion criteria. Overall, LLLT was superior to sham LLLT for relieving pain (SMD = 0.96, 95% CI 0.31–1.61) but not for improving function (SMD = 0.21, 95% CI − 0.11 to 0.53) or stiffness (SMD = 0.07, 95% CI − 0.25 to 0.39). Surface under the cumulative ranking curve (SUCRA) value ranking showed the most effective wavelength of LLLT in reducing KOA pain was 904–905 nm (SUCRA, 86.90%), followed by multi-wavelengths (MWL) (SUCRA, 56.43%) and 785–850 nm (SUCRA, 54.97%). Compared to sham LLLT, L2 (SMD = 1.42, 95% CI = 0.31–2.53) and L1 (SMD = 0.82; 95% CI = 0.11–1.50) showed a significant reduction in KOA pain. However, MWL (SMD = 0.83; 95% CI = − 0.06 to 1.72) showed similar KOA pain reduction compared to sham LLLT. The certainty of evidence showed that the quality of evidence regarding the effectiveness of overall LLLT versus sham, and 904–905 nm versus sham were low, while the quality of evidence for MWL versus sham, and 785–850 nm versus sham was very low.

Conclusion
While the 904–905 nm wavelength showed potential benefits in reducing KOA pain, the overall quality of the evidence was low. LLLT with 904–905 nm or 785–850 nm wavelengths yielded significantly better reduction in KOA pain compared to sham LLLT, but further high-quality research is warranted to validate these findings.

Full text

Vol.:(0123456789)
Aging Clinical and Experimental Research (2024) 36:203
https://doi.org/10.1007/s40520-024-02853-0
REVIEW PAPER
A systematic review andnetwork meta‑analysis ontheoptimal
wavelength oflow‑level light therapy (LLLT) intreating knee
osteoarthritis symptoms
TianxiangFan1· YangLi2· ArnoldY.L.Wong1,3· XiaoLiang1· YarouYuan1· PengXia1,4· ZhiYao5· DeliWang5·
MarcoY.C.Pang1· ChanghaiDing2· ZhaohuaZhu2· YeLi1· SiuNgorFu1
Received: 15 March 2024 / Accepted: 16 September 2024
© The Author(s) 2024
Abstract
Objectives To compare the efficacy of the various wavelengths of low-level light therapy (LLLT) in alleviating knee pain,
dysfunction, and stiffness in patients with knee osteoarthritis (KOA), and to compare the effectiveness of LLLT versus sham
treatment in reducing knee pain, dysfunction, and stiffness.
Methods PubMed, Web of Science, EMBASE, and Cochrane Library were searched from inception to 12 December 2023.
Randomized controlled trials that assessed the effects of different wavelengths of LLLT on alleviating pain of patients with
KOA were included. A conventional meta-analysis and network meta-analysis were preformed, and standardized mean dif-
ferences (SMD) with 95% confidence interval (CI) were calculated.
Results Thirteen studies involving 673 participants with KOA met inclusion criteria. Overall, LLLT was superior to sham
LLLT for relieving pain (SMD = 0.96, 95% CI 0.31–1.61) but not for improving function (SMD = 0.21, 95% CI −0.11 to
0.53) or stiffness (SMD = 0.07, 95% CI −0.25 to 0.39). Surface under the cumulative ranking curve (SUCRA) value ranking
showed the most effective wavelength of LLLT in reducing KOA pain was 904–905nm (SUCRA, 86.90%), followed by multi-
wavelengths (MWL) (SUCRA, 56.43%) and 785–850nm (SUCRA, 54.97%). Compared to sham LLLT, L2 (SMD = 1.42,
95% CI = 0.31–2.53) and L1 (SMD = 0.82; 95% CI = 0.11–1.50) showed a significant reduction in KOA pain. However,
MWL (SMD = 0.83; 95% CI = −0.06 to 1.72) showed similar KOA pain reduction compared to sham LLLT. The certainty
of evidence showed that the quality of evidence regarding the effectiveness of overall LLLT versus sham, and 904–905nm
versus sham were low, while the quality of evidence for MWL versus sham, and 785–850nm versus sham was very low.
Conclusion While the 904–905nm wavelength showed potential benefits in reducing KOA pain, the overall quality of the
evidence was low. LLLT with 904–905nm or 785–850nm wavelengths yielded significantly better reduction in KOA pain
compared to sham LLLT, but further high-quality research is warranted to validate these findings.
Keywords Low-level light therapy· Osteoarthritis· Knee pain· Meta-analysis
Tianxiang Fan and Yang Li contributed equally to this work.
* Siu Ngor Fu
amy.fu@polyu.edu.hk
1 Department ofRehabilitation Sciences, The Hong Kong
Polytechnic University, Kowloon, HongKongSAR, China
2 Clinical Research Centre, Zhujiang Hospital, Southern
Medical University, Guangzhou, Guangdong, China
3 Research Institute forSmart Ageing, The Hong Kong
Polytechnic University, Kowloon, HongKongSAR, China
4 Department ofRehabilitation Medicine, Nanjing First
Hospital, Nanjing Medical University, Nanjing, Jiangsu,
China
5 Department ofBone andJoint Surgery, Peking University
Shenzhen Hospital, Shenzhen, Guangdong, China
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 2 of 14
Introduction
Knee osteoarthritis (KOA) is a highly prevalent chronic
joint disease affecting over 654 million people world-
wide, incurring enormous societal costs, particularly
among older adults[1, 2]. KOA commonly manifests as
joint pain, stiffness, and functional limitations, leading to
reduced mobility and quality of life [3–5]. With an aging
and increasingly obese population, the prevalence of KOA
is expected to rise [3].
Low-level light therapy (LLLT) is considered a poten-
tial non-pharmaceutical therapy for knee OA. LLLT uses
laser and/or light-emitting diode (LED) with specific
wavelengths of light to stimulate cellular processes, modu-
late inflammation, and promote tissue healing [6–9]. Given
the non-invasiveness, rapid pain relief, and minimal side
effects of LLLT, it is widely used to treat various mus-
culoskeletal disorders [6]. LLLT has shown promise as a
therapeutic intervention for OA by modulating inflamma-
tion, suppressing the expression of pain-associated mol-
ecules, and improving pain behavior in animal models
of OA [10–13]. Conflicting findings have been reported
in clinical trials regarding the effectiveness of LLLT in
treating OA [14–16]. Some trials demonstrated that LLLT
was significantly better than sham LLLT in relieving pain
[17–19], while others found no significant pain reduction
[20, 21]. Similarly, systematic reviews also revealed mixed
results regarding the efficacy of LLLT plus exercise ther-
apy in alleviating pain in individuals with Knee OA (KOA)
[14, 22, 23]. There is also no consensus on whether stan-
dalone LLLT can improve OA-related pain [14, 24]. These
inconsistencies may be attributed to variations in treatment
parameters, such as wavelengths, dosages, and duration of
LLLT, as well as methodological limitations of previous
studies. Although a systematic review and meta-analysis
conducted in 2015 reported no significant superiority of
LLLT over sham LLLT in improving visual analogue scale
(VAS) pain, Western Ontario and McMaster Universities
Arthritis Index scores (WOMAC) pain, WOMAC stiffness,
or WOMAC function, this review had several limitations
[14]. It analyzed VAS pain scales and WOMAC pain scale
separately, reducing the sample size and increasing the risk
of false-negative results, and excluded some KOA studies
using alternative pain scales. Given these limitations and
the publication of seven new clinical trials since 2015,
there is a clear need for an updated systematic review and
meta-analysis to comprehensively summarize the current
evidence in this field.
The effectiveness of LLLT depends on multiple fac-
tors, including wavelengths of light, energy density, and/
or total energy. Given the diverse settings of various
treatment parameters in prior studies, they might have
confounded the findings and interpretation of the effec-
tiveness of LLLT [25]. It is noteworthy that the wavelength
of light not only affects the penetration depths of tissues
but also plays a critical role in eliciting different biologi-
cal effects [6, 26]. Some randomized control trials (RCTs)
and a meta-analysis have shown that different wavelengths
of LLLT lead to differential clinical outcomes. Specifi-
cally, Jankaew etal. found that the 808nm wavelength
group showed significantly better results in knee extensor
strength compared to the 660nm group in patients with
KOA [27]. Additionally, Ahmad etal. reported that the
1064nm LLLT combined with exercise led to significantly
better improvements in pain, physical function, and knee-
related disability than the 830nm LLLT combined with
exercise. Furthermore, a meta-analysis found that wave-
lengths of 785–860nm or 904nm significantly alleviated
pain and disability in patients with KOA [28]. However, no
network meta-analysis has compared the relative effects of
different wavelengths of LLLT on pain, stiffness, and func-
tion in patients with KOA. Therefore, a better understand-
ing of the specific wavelength range for yielding optimal
clinical outcomes is crucial for effective LLLT treatments.
The World Association for Photobiomodulation Therapy
(WALT) recommends using LLLT wavelength range of
780–860nm and 904nm for treating musculoskeletal dis-
orders [29]. However, the relative efficacy of these two
wavelength ranges in treating OA symptoms is unclear.
Further, compared to LLLT of a single radiation, a com-
bination of LLLT of different wavelengths may generate
distinct effects on different biological tissues. However,
no prior meta-analysis has compared the relative efficacy
of multi-wavelengths (MWL) of LLLT, and various sepa-
rate wavelengths of LLLT in reducing OA symptoms. This
research gap can be addressed by evaluating the efficacy
of MWL LLLT and single-wavelength LLLT through a
network meta-analysis that compares multiple interven-
tions using direct and indirect comparisons to help rank
the comparative effectiveness of various treatments [30].
Given the above, the primary objective of the current sys-
tematic review and meta-analysis was to summarize the effi-
cacy of LLLT compared to sham LLLT in improving knee
pain, stiffness and function in patients with KOA. The sec-
ondary objective was to conduct a network meta-analysis to
compare the efficacy of different wavelengths and MWL in
reducing pain, stiffness, and improving function in patients
with KOA, to identify the most effective wavelength range.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 3 of 14 203
Methods
Protocol andregistration
This systematic review and network meta-analysis was
registered with PROSPERO (ID: CRD42023396103). The
reporting of this review followed the Preferred Report-
ing Items for Systematic Reviews and Meta-Analyses
guidelines.
Search strategies andselection criteria
Four electronic databases (PubMed, Web of Science,
EMBASE, and Cochrane Library) were searched from
inception to 12 December 2023. The detailed search strat-
egies are shown in Supplementary Table1.
The following inclusion criteria were adopted based
on the PICOS framework: (1) Study design: Randomized
control trials (RCTs); (2) Patients: Patients with knee OA
on either side; (3) Intervention: LLLT was found in at least
one of the treatment groups; (4) Comparator: The interven-
tion in the control group might use sham LLLT or another
wavelength of LLLT; and (5) Outcomes: Pain intensity as
evaluated by the WOMAC, VAS, Numeric Pain Rating
Scale (NPRS), or Visual Numerical Scale (VNS) was set
as primary outcome. WOMAC knee function and stiffness
scores were set as the secondary outcomes.
Exclusion criteria: (1) missing LLLT wavelength; (2)
non-English language articles; (3) LLLT not using laser
or LED; (4) Review articles or meta-analyses; (5) stud-
ies without a control group; (6) animal or cell studies;
(7) unrelated to KOA; (8) the involvement of other thera-
pies adjunctive to LLLT; (9) no reporting of knee pain,
knee physical function or knee stiffness; (10) abstracts,
conference proceedings, grey literature or studies without
extractable data.
Identified citations were imported to EndNote 20. Two
independent reviewers (TF and YY) screened potential
titles and abstracts based on the selection criteria. Any
disagreements were discussed with a third reviewer (YL).
Eligible abstracts were retrieved for full-text screening.
The same procedure was applied for the full-text screen-
ing. Relevant full texts were included. Excellent inter-
reviewer reliability of the screening was noted (kappa
coefficient = 0.95). The reference lists of the included
studies were screened for relevant articles. Forward cita-
tion tracking was conducted using the Web of Science.
The corresponding authors of the included studies were
contacted by emails to seek pertinent articles.
Data extraction
Two independent reviewers (TF and YY) extracted data
The extracted data included authors, year of publication,
study design, sample size, age, body mass index, and
gender. Additionally, details of LLLT treatment such as
wavelengths, total energy, energy density, follow-up time,
and treatment frequency were recorded. Clinical outcomes
comprised the baseline mean, baseline standard deviation,
follow-up mean, and follow-up standard deviation for pain,
physical function, or stiffness. If standard deviations were
not reported, we extracted 95% confidence intervals (CIs),
standard errors (SEMs), or data from error bar graphs.
When only graphical data were available, numerical
data were extracted using the Engauge Digitizer 12.1
software. Any disagreements between reviewers were
resolved through discussion and consensus; if needed, a
third independent reviewer (YL) made the final decision.
To ensure data accuracy and completeness, quality checks
were conducted on a random subset of the extracted data
by an independent reviewer (YL). Based on the interven-
tion wavelengths recommended by WALT, LLLT wave-
lengths were divided into two categories: 785–860nm
(L1) and 904nm (L2). Since a study used the wavelength
of 905nm, which is very close to 904nm, the findings
were considered as the same group (904nm-905nm, L2).
The treatment group that used more than one wavelength
of light was categorized as the multi-wavelength (MWL)
group.
Pain intensity evaluated by WOMAC and/or visual pain
scale (VAS, NPRS, or VNS), WOMAC knee function and
stiffness scores were the potential clinical outcomes. In the
assessment of pain within the same study, when various pain
scales were used, we prioritized the analysis of pain scale
using the following sequence: the WOMAC pain scale, the
VAS pain scale, the NPRS pain scale, and finally, the VNS
pain scale [31].
Reviewers calculated the effect sizes by measuring the
mean difference (MD) and standard division (SD). A study
compared the effects of LLLT and sham LLLT on pain at
different time points. The findings revealed a significant
difference in pain pressure threshold at the 8th week, sug-
gesting that 8weeks may be suitable time for LLLT inter-
vention [32]. For pain, physical function, and stiffness, the
time point was at or nearest to 8weeks after initial LLLT or
sham LLLT [33].
Specifically, we extracted the original MDs and SDs in
each included study, if available. When only standard errors
(SEs) were reported, we converted them to SDs according
to the Cochrane Handbook for Systematic Reviews [34]. For
graphical information, numerical data was extracted using
the Engauge Digitizer 12.1 software (Mark Mitchell, Palos
Verdes Peninsula, CA, USA) [35]. When only the baseline
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 4 of 14
and follow-up data rather than the MDs and SDs were availa-
ble, we calculate the MDs by subtracting the follow-up mean
from the baseline mean. For studies that did not provide the
SDs of the outcome changes, the SDs were estimated using
following equation with a correlation coefficient (r) of 0.5:
according to the Cochrane Handbook guideline [34]. If none
of these options were feasible, the corresponding authors of
the included study were contacted by emails for a maximum
of three times.
Quality assessments
Two independent reviewers (TF and YL) used the Cochrane
risk-of-bias tool for randomized trials (RoB 2) to evaluate
the methodological quality of the included studies by [36].
The tool evaluates six domains: (1) randomization process;
(2) deviation from intended interventions; (3) missing out-
come data; (4) measurements of outcomes; (5) selection
of the reported results; and (6) overall bias. Each domain
was classified as low, some concerns, or high risk of bias
based on criteria from the Cochrane Handbook for System-
atic Reviews [36]. Discrepancies between reviewers were
resolved through discussion and consensus. If consensus
could not be reached, the final decision was made after dis-
cussion with the third reviewer (YY). The overall bias of
each study was corresponds to the worst risk of in any of the
first five domains, as suggested by the BMJ guideline [36].
Certainty oftheevidence
The grading of recommendation assessment, development,
and evaluation (GRADE) approach was utilized in this meta-
analysis to assess the certainty of evidence [37]. Two inde-
pendent reviewers (TF and PX) conducted assessments for
each comparison and resolved any discrepancies through
consensus. Certainty ratings were assigned for each com-
parison and endpoint, with ratings of high, moderate, low,
or very low, based on a thorough evaluation of risk of bias,
inconsistency, indirectness, publication bias, intransitivity,
incoherence and imprecision according to the GRADE hand-
book [38].
Statistical analyses
All the calculations and figures drawing were conducted in R
software, version 4.1.3. Conventional meta-analyses compar-
ing LLLT with sham LLLT were conducted for each KOA
symptomatic outcomes using “meta” and “metafor” pack-
ages [39, 40]. The “meta” package’s “metacont” function
SD
change=
√
SD2
baseline+SD2
follow−up−(2r×SDbaseline ×SDfollow−up
was used to calculate common and random effects estimates
for meta-analyses with continuous outcome data, utilizing
inverse variance weighting for pooling. The ‘forest’ function
from the “metafor” package generated forest plots based on
results calculated by “metacont”. A random effects model
network meta-analyses were then conducted to explore the
relative efficacy of LLLT with different wavelengths, using
the “netmeta” package based on the frequentist framework
[41]. The consistency between direct and indirect compari-
sons was tested by node-splitting analysis.
The I2 test was conducted to assess the heterogeneity of
each pairwise comparisons by “metacont”, with I2 exceed-
ing 50% indicating heterogeneity[42]. Because the included
studies used slightly different scales for measuring various
clinical outcomes, changes from baseline to follow-up were
converted to standardized mean difference (SMD). SMDs
were calculated as the difference in mean outcomes between
groups divided by the standard deviation of the outcome
among participants [34]. Regarding the calculation of stand-
ard deviations, we employed the pooled standard deviation
method.We first calculated the baseline and follow-up
standard deviations for both the experimental and control
groups. Then, the standard deviation change in each group
was incorporated into the “metacont” function in R for fur-
ther calculation using a pooled standard deviation.
To numerically rank the associations between all inter-
ventions and pain reduction, the surface under the cumula-
tive ranking curve (SUCRA) was calculated [43]. The value
of SUCRA ranges from 0–100%. The SUCRA values rep-
resent the probability of an intervention being ranked as the
best, second best, and so on, for the outcome of interest.
A SUCRA value of 100% indicates that the intervention is
certain to be the most effective, while a value of 0% indicates
that it is certain to be the least effective.
To assess publication bias, the Egger regression test and
the funnel plot were employed using the “meta” package. In
this analysis, a p-value less than 0.10 was considered signifi-
cant, indicating the presence of asymmetry and publication
bias [44]. Sensitivity analysis was conducted by excluding
studies with high risk of overall bias. Two-sided tests were
conducted for all analyses, and statistical significance was
defined as a p-value less than 0.05.
Results
Study selection
A total of 818 records were identified in the initial search
from four databases and 199 records from the forward cita-
tion tracking in Web of Science. Figure1 illustrates the
screening procedure and the number of articles that met
the inclusion criteria. After removing 439 duplicates, 578
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 5 of 14 203
articles were screened for titles and abstracts. Finally, a
total of 13 RCTs were included in the analysis (Fig.1)
[17, 19–21, 45–53]. A total of 12 studies employed lasers
as the light source for LLLT, while one study conducted
by Nathali Cordeiro Pinto etal. utilized LED light source
[47].
Study characteristics
Table1 shows the demographic characteristics of the
included studies. A total of 673 participants from 13 tri-
als were included in the analysis, the sample size of these
trials ranged from 29 to 101. The mean age of included
patients ranged from 55.2 to 69.0years. All participants
were categorized into four groups according to their
received treatments: 785–850nm LLLT (L1), 904–905nm
LLLT (L2), multi-wavelength LLLT group (MWL), sham
LLLT (sham). All included studies evaluated knee pain.
Four included studies assessed knee physical function
and stiffness. Eight included studies were double-blinded
RCTs, while five trials were single-blinded RCTs.
Risk ofbias assessments
Figure2 depicts the risk of bias assessments of the included
studies. Two studies had a high rate of missing values, which
skewed the results due to inadequate analysis methods.
Another study was deemed high risk because participants
may have been aware of their assigned intervention during
the trial. Additionally, five studies were rated as having some
concerns. These studies had some concerns with deviations
from intended interventions, selective reporting of results, or
some missing outcome data. Although these concerns were
not severe enough to classify the studies as high risk, they
indicated minor inconsistencies in intervention administration,
selective outcome reporting, and inadequate handling of miss-
ing data. Four studies demonstrated a low risk of bias. The
risk-of-bias summary and graph are shown in Fig.2.
Fig. 1 A flow diagram of the study identification and selection process
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 6 of 14
Table 1 Characteristics of the included studies
First author,
Publication
year
Study
Design
Sample
size
Age (years) Female
(%)
BMI Follow-
up
(week)
Fre-
quency,
duration/
session,
number of
sessions
Wavelength, type
of device
Treatment
area
Total
energy/
session
Energy
density (J/
cm2)
Primary
outcome
Secondary
outcome
Punpetch
Siriratna,
2022 [48]
SB-RCT 21 66.1 ± 9.4 85.7 28.1 ± 5.2 4 2–3 ses-
sions/
week for
a total,
8min/
session,
10 ses-
sions
808nm + 905nm,
laser
8 points
around
the knee
joint
562.50 22.39 VAS pain –
21 65.0 ± 8.5 76.2 27.4 ± 5.8 0 0 0 –
Renata
Alqualo-
Costa, 2021
[49]
DB-RCT 42 61.3 ± 9.4 83% 31.2 ± 6.1 4 3 sessions/
week,
75s/ses-
sion, 12
sessions
904nm, laser Pain areas 27 54 VAS pain –
42 65.3 ± 8.5 78% 29.9 ± 4.6 0 0 0
Patricia Gabri-
elli Vassão,
2020 [50]
DB-RCT 14 64.00 ± 4.93 –29.08 ± 4.81 8 2 sessions/
week,
40s/ses-
sion, 16
sessions
808nm, laser 7 points
around
knee
56 91 NPRS
pain
–
15 65.37 ± 4.19 –27.52 ± 3.31 0 0 0
Roberta de
Matos
Brunelli
Braghin,
2019 [20]
DB-RCT 15 58.20 ± 7.97 87% 31.57 ± 3.58 8 2 sessions/
week,
56s/ses-
sion, 15
sessions
808nm, laser 10 points
around
knee
200 56 WOMAC
pain
WOMAC
stiffness,
WOMAC
physical
function
15 60.8 ± 9.2 80% 26.52 ± 4.43 0 0 0
Gopal Nambi
S, 2017 [14]
DB-RCT 17 58 ± 6 –26.9 ± 4.8 4 3 sessions/
week,
60s/ses-
sion, 12
sessions
905nm, laser 8 points
around
the knee
12 1.5 VAS pain –
17 60 ± 8 –28.3 ± 3.5 0 0 0
Ahmad
Alghadir,
2014 [18]
SB-RCT 20 55.2 ± 8.14 50% 32.34 ± 5.77 4 2 sessions/
week,
10min/
session,
8 ses-
sions
850nm, laser 8 points
around
the joint
48 48 WOMAC
pain
WOMAC
stiffness,
WOMAC
physical
function
20 57 ± 7.77 40% 33.09 ± 4.98 0 0 0
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 7 of 14 203
Table 1 (continued)
First author,
Publication
year
Study
Design
Sample
size
Age (years) Female
(%)
BMI Follow-
up
(week)
Fre-
quency,
duration/
session,
number of
sessions
Wavelength, type
of device
Treatment
area
Total
energy/
session
Energy
density (J/
cm2)
Primary
outcome
Secondary
outcome
Nelson Mar-
quina, 2012
[19]
SB-RCT 53 - 41.50% – 6 3 sessions/
week,
1min/
session,
12 ses-
sions
905nm + 660, laser
nm
7 points
around
the joint
1.68 24 VAS pain –
48 - 31.25% – 0 0 0
Kamila Gwo-
rys, 2012 [52]
DB-RCT 31 67.7 ± 11.3 – – 2 5 sessions/
week,-,
10 ses-
sions
0 0 0 VAS pain –
34 57.6 ± 11.8 – – 810nm, laser 7 points
around
the joint
152.4 12.7
31 65.4 ± 9.6 – – 808nm + 905, laser
nm
148.8 6.21
Funda Tas-
cioglu, 2004
[21]
SB-RCT 20 62.86 ± 7.32 70% 27.56 ± 5.65 3 5 sessions/
week,
10min/
session,
10 ses-
sions
830nm, laser 5 points
around
the joint
15 382 WOMAC
pain
WOMAC
stiffness,
WOMAC
physical
function
20 64.27 ± 10.55 65% 29.56 ± 9.54 0 0 0
VanessaOva-
nes-
sianFukuda,
2011 [53]
DB-RCT 25 63.0 ± 9.0 20% 30.0 ± 3.5 3 3 sessions/
week,
7.5min/
session,
9 ses-
sions
904nm, laser 9 points
around
the joint
27 6 VNPS
pain
-
22 63.0 ± 8.0 36% 28.7 ± 4.1 0 0 0
XueyongShen,
2009 [45]
SB-RCT 20 60.10 ± 6.83 10% – 4 3 sessions/
week,
20min/
session,
12 ses-
sion
650nm + 1060nm,
laser
Acupunc-
ture point
Dubi (ST
35)
- - WOMAC
pain
WOMAC
stiffness,
WOMAC
physical
function
20 56.40 ± 7.41 10% – 0 0 0
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 8 of 14
Table 1 (continued)
First author,
Publication
year
Study
Design
Sample
size
Age (years) Female
(%)
BMI Follow-
up
(week)
Fre-
quency,
duration/
session,
number of
sessions
Wavelength, type
of device
Treatment
area
Total
energy/
session
Energy
density (J/
cm2)
Primary
outcome
Secondary
outcome
DwiR.Helian-
thi, 2016 [46]
DB-RCT 30 69 ± 6.0 60% 25.8 ± 4.3 6.5 2 sessions/
week,
5min
20s/ses-
sion, 10
sessions
785nm, laser Acupunc-
ture
points
of ST35
Dubi,
ST36
Zusanli,
SP9 Yin-
lingquan,
GB34
Yangling-
quan and
EX-LE-4
Neixiyan
20 10 VAS pain –
29 68 ± 5.0 60% 26.3 ± 4.3 0 0 0
Nathali Cord-
eiro Pinto,
2022 [47]
DB-RCT 15 63 ± 10.9 93% 24.8 ± 9.6 8 2 sessions/
week,
5–8min/
session,
10 ses-
sions
850nm, light-emit-
ting diode
Whole
joint
surface
526–1402 18–48 NPRS
pain
–
16 66 ± 10.7 94% 29.8 ± 4.6 0 0
DB-RCT Double blind-randomized controlled trial, SB-RCT Single blind-randomized controlled trial, BMI Body mass index, WOMAC Western Ontario and McMaster Universities Arthritis
Index scores, VAS Visual analogue scale, NPRS Numeric pain rating Scale, VNS Visual numerical scale
Mean (standard deviation) are provided above for age, BMI
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 9 of 14 203
Conventional meta‑analyses oftheeffects ofLLLT
interventions onknee pain, physical function,
andstiffness
Findings from 13 included studies showed that LLLT was
superior to sham LLLT for pain relief (SMD = 0.96, 95% CI
0.31–1.61) (Fig.3). There is low certainty of evidence sup-
porting overall LLLT for knee pain, primarily due to one
trial showing a high risk of bias. Additionally, significant
heterogeneity across 13 studies was found in meta-analy-
sis, as indicated by an I2 value of 86%. Similar concerns
apply to the effectiveness of 904–905nm LLLT, with an I2
of 95%. Furthermore, findings from network meta-analysis
also decreased the certainty of evidence. Evidence for the
effectiveness of multiwavelength LLLT was very low, as
indicated by the 95% CI of SMD including zero, alongside
results from network meta-analysis. Additionally, there is
very low certainty in the effectiveness of 785–850nm LLLT
for knee pain, compounded by missing data in one trial,
an I2 of 88.8%, and evidence from network meta-analysis.
(Supplementary Table2 & Supplementary Fig.1). Addi-
tionally, the results from four included studies showed that
LLLT was not significantly better than sham LLLT for knee
function (SMD = 0.21, 95% CI −0.11 to 0.53) or knee stiff-
ness (SMD = 0.07, 95% CI −0.25 to 0.39) (Supplemen-
tary Figs.6 and 7). The results of the sensitivity analysis
that excluded high-risk studies for knee pain consistently
showed that LLLT was superior to sham LLLT for pain relief
(SMD = 1.13, 95% CI 0.15–2.12) (Supplementary Fig.4).
Network meta‑analysis oftheeffects ofLLLT
interventions onknee pain, physical function,
andstiffness
Thirteen studies were included in the network meta-
analysis to compare the relative effectiveness of different
wavelengths of LLLT and sham LLLT in improving knee
pain. Compared with sham LLLT, L2 (SMD = 1.42, 95%
CI = 0.31–2.53) and L1 (SMD = 0.82; 95% CI = 0.11–1.50)
Fig. 2 Risk of bias assessments. A the judgement of each bias item
for each included study. B judgements of each bias item presented as
percentages across all included studies
Fig. 3 A forest plot of evidence from the direct comparisons between LLLT and sham LLLT for knee pain. SMD standardized mean difference,
SD standard division
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 10 of 14
significantly improved knee pain, while MWL (SMD = 0.83;
95% CI = −0.06 to 1.72) did not show a significant effect
(Table2). However, L2 was not superior to L1 or MWL in
relieving knee pain. The rank plot recommends L2 (SUCRA,
86.70%) seems to be the most effective therapy for knee
pain relief, followed by MWL (SUCRA, 57.01%) and L1
(SUCRA, 54.97%) (Fig.4). The results of the sensitivity
analysis after excluding high-risk studies showed that only
L2 (SMD = 1.56, 95% CI = 0.02–3.09) was effective in
relieving knee pain (Supplementary Fig.5). Node-splitting
analysis indicated a good global and local consistency of the
network for knee pain (Supplementary Table2).
For the included studies that assessed knee physical func-
tion and stiffness, only L1, MWL or sham LLLT were used
as interventions. Neither L1 nor MWL showed significant
improvement in physical function or knee stiffness compared
to sham LLLT (Supplementary Figs.6 & 7).
Discussion
This systematic review and network meta-analysis is an
update of a previous systematic review, and found low-cer-
tainty evidence to support LLLT was effective for alleviating
KOA pain at eight weeks post-treatment. Specifically, LLLT
with both L1 and L2 wavelength was superior to sham LLLT
in knee pain reduction. According to SUCRA ranking, L2
was the optimal wavelength of LLLT in reducing knee pain,
followed by MWL and L1. LLLT of any wavelength was
found no effect on KOA function and stiffness.
Several previous meta-analyses revealed inconsistent
results regarding the effects of LLLT on KOA. A meta-
analysis by Huang etal. found no significant effect of LLLT
on knee pain, function, or stiffness [23]. Their search was
out of dated (conducted between January 2000 and Novem-
ber 2014), resulting in only nine included studies with rela-
tively small sample sizes [23]. Conversely, a meta-analysis
including 14 studies found that LLLT significantly improved
pain, function, and stiffness in individuals with KOA [23].
However, this meta-analysis did not identify the detailed
Table 2 Network meta-analysis of different wavelengths of LLLT
with Surface under the cumulative ranking curve (SUCRA) value for
knee pain
League tables showing the results of the network meta-analyses, with
the different standardized mean difference (SMD) and 95% credible
intervals in the lower left part of the table, and the SUCRA values
presented in the upper right part
Bold numbers are statistically significant
LLLT low-level light therapy, L1 785–850nm LLLT, L2 904–905nm
LLLT, MWL multi-wavelength LLLT, Sham sham LLLT
SUCRA value
L1 86.90%
−0.61 (−1.92,
0.70); L2 54.97%
−0.02 (−1.08,
1.04)
0.59 (−0.83, 2.01) MWL 56.43%
0.81 (0.11, 1.50) 1.42 (0.31, 2.53) 0.83
(−0.06,
1.72)
Sham LLLT
Fig. 4 A network plot and Surface under the cumulative ranking
curve (SUCRA) plot for knee pain. A Different nodes represent dif-
ferent intervention groups. The size of the nodes is proportional to the
number of patients who were assigned to the intervention. The thick-
ness of the lines connecting the nodes is proportional to the number
of pairwise trials that evaluated the interventions. B SUCRA plot. L2
has the largest probability to be the first best treatment for knee pain.
L1 785–850 nm LLLT, L2 904–905nm LLLT, MWL multi-wave-
length LLLT, Sham sham LLLT, LLLT low-level light therapy
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 11 of 14 203
parameters of LLLT (e.g., wavelengths), leading to signifi-
cant heterogeneity in the result [23]. Different wavelengths
of LLLT not only exhibit varying tissue penetration capabili-
ties, but also induce different biological effects [54, 55]. The
inconsistent findings across different studies investigating
LLLT for treating KOA might be attributed to the varia-
tions in LLLT parameters [56]. A Cochrane Library review
highlighted the importance of consistent reporting of the
characteristics of LLLT devices [56]. Despite the positive
findings of some studies, previous meta-analyses lacked data
on how the effectiveness of LLLT is influenced by various
important factors, including wavelength, duration, dosage,
and treatment site [56]. Prior to this review, the optimal
treatment parameters for KOA using LLLT were unclear.
Although the WALT gave recommendations on the wave-
length for phototherapy, there was insufficient evidence to
support their suggested wavelength for OA [29]. Our analy-
sis revealed a significant reduction in KOA pain using LLLT
with wavelength of 785–850nm or 904–905nm. Accord-
ing to SUCRA ranking, 904–905 was the optimal wave-
length of LLLT in reducing knee pain, followed by MWL
and L1. However, it is important to note that the confidence
intervals for the SMDs overlap considerably, indicating a
degree of uncertainty in the comparative effectiveness of
these wavelengths. Further research is needed to confirm
these findings and to establish the optimal wavelengths for
LLLT in the treatment of KOA. Additionally, we did not
observe significant pain reduction with MWL. Our findings
provide low-certainty evidence to support the selection of
optimal wavelength for future LLLT treatment of KOA. A
prior systematic review found that LLLT was beneficial as
an adjunct treatment to exercise in Incoherence managing
KOA[57]. However, our review excluded studies containing
treatment groups that combined exercise therapy with LLLT
due to heterogeneous exercise protocols. The heterogeneity
in exercise protocols could substantially impact the model's
transitivity in a network meta-analysis, leading to unreliable
results in indirect comparisons [34]. Transitivity is a crucial
assumption, indicating that included studies should be simi-
lar in their distributions of effect modifiers. Violating this
assumption can result in misleading indirect comparisons.
Differences in exercise protocols across studies can intro-
duce variability that affects the comparability of interven-
tions, potentially biasing estimates of treatment effects and
undermining the reliability of conclusions. The observed
superior effectiveness of L2 compared to L1 in reducing pain
may be attributed to the potential mechanism involving the
stronger penetration of LLLT with wavelengths 904–905nm
compared to 785–850nm [58]. The 904-905nm wavelength
of LLLT, known for its deeper tissue penetration, may ben-
efit patients with severe cartilage damage and obesity-related
KOA. Its enhanced penetration is crucial for treating deep
joint cartilage damage effectively, making it particularly
advantageous for obese patients [59].
Although LLLT with wavelengths 785–850nm may
reduce inflammation and pain [60, 61], LLLT at such wave-
lengths has difficulty in penetrating the knee joint. This may
lead to limited biological responses due to absorption by
the joint tissues, affecting the efficacy. Furthermore, multi-
wavelength LLLT did not show significant difference in pain
reduction as compared to sham LLLT. One possible rea-
son is that different studies used different combinations of
wavelengths as their multi-wavelength LLLT, which might
confound the findings. Another possible reason could be that
some studies in the multi-wavelength LLLT group used very
low energy or energy density [19, 52], which may not be
sufficient to induce significant biological effects. Interest-
ingly, despite the larger SMD favoring L2 over L1, there
was no significant difference between these two groups. This
may be attributed to the fact that the comparisons between
L2 and L1 were based on indirect evidence, resulting in a
high degree of incoherence and imprecision in the results.
Regarding LLLT on stiffness and physical function, both the
conventional meta-analysis and network meta-analysis did
not find a significant effect compared to sham LLLT. This
could be due to the absence of studies specifically investi-
gating stiffness and physical function using the 904–905nm
wavelength. Further clinical trials are warranted to explore
these aspects in the future. The inclusion of a multi-arm
study included into our network meta-analysis, which could
impact results in several ways. Firstly, a multi-arm study pro-
vides direct comparative data between two treatment intensi-
ties, increasing the number of direct comparisons within the
analysis and potentially enhancing the statistical power and
certainty of evidence for those treatments [62]. Secondly, a
multi-arm study enhances network connectivity by linking
treatments that might only be indirectly connected through
other paths, which could improve the accuracy and stability
of overall network estimates [63]. Finally, the likelihood of
publication bias may increase if multi-arm studies, perceived
as more comprehensive or informative, are more likely to
be published, especially if they report more favorable out-
comes, potentially influencing the direction of the overall
conclusions.
The current review has several strengths. First, the current
protocol was registered with PROSPERO. Second, standard-
ized procedures were conducted and the reporting followed
the PRISMA guidelines. Third, this is the first network
meta-analysis comparing the efficacy of LLLT with differ-
ent wavelengths for treating KOA symptoms. We utilized
reliable methods recommended by the Cochrane Collabora-
tion for the network meta-analysis. Moreover, we employed
SUCRA outcomes ratings to identify any subtle differences
among these treatments.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 12 of 14
This review has several limitations. First, only 13 small-
scale studies were included in our analysis. Specifically, the
comparison between MWL and L1 was based on one sin-
gle trial with 65 patients. This limited sample size affected
the generalizability of our findings and introduced potential
bias. Second, only four included studies evaluated the effects
of LLLT on knee physical activity and stiffness. The small
sample size and the use of LLLT with wavelength longer
than 904–905nm might lead to non-significant results.
Third, the current review only considered the effects of
LLLT wavelength on KOA outcomes at the 8-week follow-
up. Because other confounders, such as LLLT frequency,
or energy density, treatment duration, treatment locations,
types of KOA, follow-up time points, may affect the results,
our findings should be interpreted with great caution. Future
studies should use standardized protocol with different dos-
ages to systematically evaluate the effects of wavelength,
frequency, and energy density of LLLT on various clini-
cal outcomes of patients with KOA in short- and long-term
follow-ups. Fourth, the SD values from one RCT included
in the analysis were derived from images through software,
which could potentially lead to inaccuracies. That said,
the estimated SD values were calculated using the official
methods outlined in the Cochrane Handbook for Systematic
Reviews of Interventions, and the image data were extracted
using the reliable Engauge Digitizer software [34]. Fifth,
four studies included in the network analysis for knee pain
had a high risk of bias. After excluding these studies in our
sensitivity analysis, 904–905nm remained to be the optimal
wavelength for KOA pain reduction; however, there was no
significant difference in pain relief between L1 and sham
LLLT. Sixth, we only extracted data at or nearest to eight
weeks for meta-analyses. Future studies should compare
the treatment effects at different follow-up time points. Sev-
enth, As the network meta-analysis was drawn from indirect
comparisons, they should be interpreted with care. Further
clinical trials should directly compare the effects of different
wavelengths on KOA symptoms. Eighth, studies investigat-
ing the combination of LLLT with exercise or other treat-
ments for knee KOA were excluded in the current review,
which might limit the generalizability of our findings. Ninth,
our meta-analysis might have been affected by publication
bias. The Egger test indicated significant publication bias
among comparisons, and funnel plots showed asymmetry,
suggesting that potential omission of smaller studies with
non-significant results from the analysis. Tenth, in this study,
the knee pain questionnaires used related to the general con-
dition of the knee rather than to particular areas of the knee
joint. Considering that knee pain can originate from multiple
tissues [64], different depths of knee pain may respond vari-
ably to various wavelengths of LLLT. Future research is also
needed to investigate the effects of different wavelengths of
LLLT on pain in specific areas of the knee joint that are less
easily treated. This would contribute to a more comprehen-
sive understanding of the applications of LLLT in targeting
different sources and locations of knee pain. Eleventh, our
meta-analysis only included studies published in English,
which may introduce language bias and limit the compre-
hensiveness of our findings.
Conclusions
Our systematic review and meta-analysis found low certainty
of evidence that LLLT could effectively reduce KOA pain,
and LLLT with wavelengths 904–905nm might be the most
effective wavelength range for relieving KOA pain. Our find-
ings provided preliminary support for using LLLT to treat
KOA pain, but not for improving knee function or stiffness.
Future studies should systematically evaluate the influences
of different treatment parameters on modifying the clinical
outcomes of patients with KOA.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s40520- 024- 02853-0.
Acknowledgements Not applicable.
Author contribution Tianxiang Fan, Siu Ngor Fu came up with this
research topic. Tianxiang Fan and Yang Li analysed the data. Tianxiang
Fan wrote the manuscript. Tianxiang Fan, Yang Li and Yarou Yuan
searched the databases and retrieved data from the included studies.
Tianxiang Fan, Yang Li and Peng Xia accessed the risk of bias or
certainty of evidence. Arnold Y L Wong, Xiao Liang, Ye Li, Zhao-
hua Zhu, Yang Li, Zhi Yao, Deli Wang, Marco YC Pang, Changhai
Ding, Siu Ngor FU provided editing and intellectual inputs. All authors
revised the manuscript for important intellectual content and approved
the final version.
Funding The present study was supported by Support related to
Research Institute for Sports Science and Technology, RISports
(P0043199).
Data Availability The original contributions presented in the study are
included in the article/supplementary material. Further inquiries can
be directed to the corresponding authors.
Declarations
Conflict of interest The authors declare that they have no competing
interests.
Ethical approval and consent to participate Not applicable.
Consent for publication Not applicable.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 Page 13 of 14 203
the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
References
1. GBD (2021) Osteoarthritis Collaborators (2023) Global, regional,
and national burden of osteoarthritis, 1990–2020 and projections
to 2050: a systematic analysis for the global burden of disease
study 2021. Lancet Rheumatol 5:e508–e522. https:// doi. org/ 10.
1016/ S2665- 9913(23) 00163-7
2. Cui A, Li H, Wang D etal (2020) Global, regional prevalence,
incidence and risk factors of knee osteoarthritis in population-
based studies. EClinicalMedicine. https:// doi. org/ 10. 1016/j.
eclinm. 2020. 100587
3. Hunter DJ, Bierma-Zeinstra S (2019) Osteoarthritis. Lancet
393:1745–1759. https:// doi. org/ 10. 1016/ S0140- 6736(19) 30417-9
4. Guo R, Ou Y-N, Hu H-Y etal (2022) The association between
osteoarthritis with risk of dementia and cognitive impairment: a
meta-analysis and systematic review. J Alzheimers Dis 89:1159–
1172. https:// doi. org/ 10. 3233/ JAD- 220568
5. Nüesch E, Dieppe P, Reichenbach S etal (2011) All cause and dis-
ease specific mortality in patients with knee or hip osteoarthritis:
population based cohort study. BMJ 342:638. https:// doi. org/ 10.
1136/ bmj. d1165
6. Dompe C, Moncrieff L, Matys J etal (2020) Photobiomodula-
tion-underlying mechanism and clinical applications. J Clin Med.
https:// doi. org/ 10. 3390/ jcm90 61724
7. He K, Zhou X, Zheng F etal (2023) Histological, physiological
and biomechanical effects of low-level laser therapy on tendon
healing in animals and humans: a systematic review. Ann Biomed
Eng 51:2659–2707. https:// doi. org/ 10. 1007/ s10439- 023- 03364-1
8. Al Zoubi FM, Wong AYL, Cheing GLY etal (2023) Adapting a
clinical practice guideline for management of patients with knee
and hip osteoarthritis by hong kong physiotherapists. Healthc.
https:// doi. org/ 10. 3390/ healt hcare 11222 964
9. Mariner Gonzalez A, Reyes SG, Ho AA etal (2024) Underrep-
resentation of non-white participants in the american academy
of orthopaedic surgeons guidelines for surgical management of
knee osteoarthritis. J Arthroplasty 39:520–526. https:// doi. org/
10. 1016/j. arth. 2023. 08. 023
10. Assis L, Milares LP, Almeida T etal (2016) Aerobic exercise
training and low-level laser therapy modulate inflammatory
response and degenerative process in an experimental model of
knee osteoarthritis in rats. Osteoarthr Cartil 24:169–177. https://
doi. org/ 10. 1016/j. joca. 2015. 07. 020
11. Martins LPO, Santos FFD, Costa TED etal (2021) Photobiomod-
ulation therapy (light-emitting diode 630nm) favored the oxida-
tive stress and the preservation of articular cartilage in an induced
knee osteoarthritis model. Photobiomodul Photomed Laser Surg
39:272–279. https:// doi. org/ 10. 1089/ photob. 2020. 4926
12. Tomazoni SS, Leal-Junior ECP, Pallotta RC etal (2017) Effects of
photobiomodulation therapy, pharmacological therapy, and physi-
cal exercise as single and/or combined treatment on the inflamma-
tory response induced by experimental osteoarthritis. Lasers Med
Sci 32:101–108. https:// doi. org/ 10. 1007/ s10103- 016- 2091-8
13. Kim G, Kim E (2013) Analgesic efficacy of low intensity laser
therapy in a monosodium iodoacetate-induced osteoarthritic rat
model. J Phys Ther Sci 25:309–312
14. Huang Z, Chen J, Ma J etal (2015) Effectiveness of low-level laser
therapy in patients with knee osteoarthritis: a systematic review
and meta-analysis. Osteoarthr Cartil 23:1437–1444. https:// doi.
org/ 10. 1016/j. joca. 2015. 04. 005
15. Chow R, Liebert A, Tilley S etal (2021) Guidelines versus
evidence: what we can learn from the Australian guideline
for low-level laser therapy in knee osteoarthritis? a narrative
review. Lasers Med Sci 36:249–258. https:// doi. org/ 10. 1007/
s10103- 020- 03112-0
16. Practitioners RAC (2018) Guideline for the management of knee
and hip osteoarthritis, 2nd edn. RACGP, Australia
17. Gopal Nambi S, Kamal W, George J etal (2017) Radiological
and biochemical effects (CTX-II, MMP-3, 8, and 13) of low-
level laser therapy (LLLT) in chronic osteoarthritis in Al-Kharj,
Saudi Arabia. Lasers Med Sci 32:297–303. https:// doi. org/ 10.
1007/ s10103- 016- 2114-5
18. Alghadir A, Omar MTA, Al-Askar AB etal (2014) Effect of
low-level laser therapy in patients with chronic knee osteoar-
thritis: a single-blinded randomized clinical study. Lasers Med
Sci 29:749–755. https:// doi. org/ 10. 1007/ s10103- 013- 1393-3
19. Marquina N, Dumoulin-White R, Mandel A etal (2012) Laser
therapy applications for osteoarthritis and chronic joint pain - a
randomized placebo-controlled clinical trial. Photonics Lasers
Med 1:299–307. https:// doi. org/ 10. 1515/ plm- 2012- 0030
20. de Matos BrunelliBraghin R, Libardi EC, Junqueira C, etal
(2019) The effect of low-level laser therapy and physical exer-
cise on pain, stiffness, function, and spatiotemporal gait vari-
ables in subjects with bilateral knee osteoarthritis: a blind ran-
domized clinical trial. Disabil Rehabil 41:3165–3172. https://
doi. org/ 10. 1080/ 09638 288. 2018. 14931 60
21. Tascioglu F, Armagan O, Tabak Y etal (2004) Low power laser
treatment in patients with knee osteoarthritis. Swiss Med Wkly
134:254–258. https:// doi. org/ 10. 4414/ smw. 2004. 10518
22. Ahmad MA, Mohamad MS, Yusof A (2022) Effects of low-level
and high-intensity laser therapy as adjunctive to rehabilitation
exercise on pain, stiffness and function in knee osteoarthritis: a
systematic review and meta-analysis. Physiother (United King-
dom) 114:85–95. https:// doi. org/ 10. 1016/j. physio. 2021. 03. 011
23 Rayegani SM, Raeissadat SA, Heidari S etal (2017) Safety and
effectiveness of low-level laser therapy in patients with knee
osteoarthritis: a systematic review and meta-analysis. J Lasers
Med. https:// doi. org/ 10. 15171/ jlms. 2017. s3
24. Stausholm MB, Bjordal JM, Lopes-Martins RAB etal (2017)
Methodological flaws in meta-analysis of low-level laser therapy
in knee osteoarthritis: a letter to the editor. Osteoarthr Cartil
25:e9–e10. https:// doi. org/ 10. 1016/j. joca. 2016. 09. 022
25. Ren H, Liu J, Liu Y etal (2022) Comparative effectiveness of
low-level laser therapy with different wavelengths and trans-
cutaneous electric nerve stimulation in the treatment of pain
caused by temporomandibular disorders: a systematic review
and network meta-analysis. J Oral Rehabil 49:138–149. https://
doi. org/ 10. 1111/ joor. 13230
26. de Freitas LF, Hamblin MR (2016) Proposed mechanisms of
photobiomodulation or low-level light therapy. IEEE J Sel Top
Quantum Electron. https:// doi. org/ 10. 1109/ JSTQE. 2016. 25612
01
27. Jankaew A, You Y-L, Yang T-H etal (2023) The effects of low-
level laser therapy on muscle strength and functional outcomes
in individuals with knee osteoarthritis: a double-blinded rand-
omized controlled trial. Sci Rep 13:165. https:// doi. org/ 10. 1038/
s41598- 022- 26553-9
28. Stausholm MB, Naterstad IF, Joensen J etal (2019) Efficacy of
low-level laser therapy on pain and disability in knee osteoarthri-
tis: systematic review and meta-analysis of randomised placebo-
controlled trials. BMJ Open 9:e031142. https:// doi. org/ 10. 1136/
bmjop en- 2019- 031142
29. Bjordal JM (2012) Low level laser therapy (LLLT) and world
association for laser therapy (WALT) dosage recommendations.
Photomed Laser Surg 30:61–62. https:// doi. org/ 10. 1089/ pho.
2012. 9893
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Aging Clinical and Experimental Research (2024) 36:203 203 Page 14 of 14
30. Dias S, Caldwell DM (2019) Network meta-analysis explained. Arch
Dis Child Fetal Neonatal Ed 104:F8–F12. https:// doi. org/ 10. 1136/
archd ischi ld- 2018- 315224
31. Beaudart C, Lengelé L, Leclercq V etal (2020) Symptomatic
efficacy of pharmacological treatments for knee osteoarthritis:
a systematic review and a network meta-analysis with a 6-month
time horizon. Drugs 80:1947–1959. https:// doi. org/ 10. 1007/
s40265- 020- 01423-8
32. Stausholm MB, Naterstad IF, Alfredo PP etal (2022) Short- and
long-term effectiveness of low-level laser therapy combined with
strength training in knee osteoarthritis: a randomized placebo-con-
trolled trial. J Clin Med. https:// doi. org/ 10. 3390/ jcm11 123446
33. Goh S-L, Persson MSM, Stocks J etal (2019) Efficacy and poten-
tial determinants of exercise therapy in knee and hip osteoarthri-
tis: a systematic review and meta-analysis. Ann Phys Rehabil Med
62:356–365. https:// doi. org/ 10. 1016/j. rehab. 2019. 04. 006
34. Higgins JPT, Thomas J, Chandler J etal (2019) Cochrane handbook
for systematic reviews of interventions. John Wiley & Sons
35. No DJ, Inkeles MS, Amin M etal (2018) Drug survival of biologic
treatments in psoriasis: a systematic review. J Dermatolog Treat
29:460–466
36. Sterne JAC, Savović J, Page MJ etal (2019) RoB 2: a revised tool for
assessing risk of bias in randomised trials. BMJ 366:l4898. https://
doi. org/ 10. 1136/ bmj. l4898
37. Puhan MA, Schünemann HJ, Murad MH etal (2014) A GRADE
Working Group approach for rating the quality of treatment effect
estimates from network meta-analysis. BMJ 349:g5630. https:// doi.
org/ 10. 1136/ bmj. g5630
38. Schünemann H, Brożek J, Guyatt G, Oxman A (2013) The GRADE
handbook
39. Schwarzer G (2007) Meta: an R package for meta-analysis. R news
7:40–45
40. Viechtbauer W (2010) Conducting meta-analyses in R with the
metafor package. J Stat Softw 36:1–48
41. Neupane B, Richer D, Bonner AJ etal (2014) Network meta-analysis
using R: a review of currently available automated packages. PLoS
ONE 9:e115065
42. Higgins JPT, Thompson SG (2002) Quantifying heterogeneity in a
meta-analysis. Stat Med 21:1539–1558. https:// doi. org/ 10. 1002/ sim.
1186
43. Mbuagbaw L, Rochwerg B, Jaeschke R etal (2017) Approaches
to interpreting and choosing the best treatments in network meta-
analyses. Syst Rev 6:79. https:// doi. org/ 10. 1186/ s13643- 017- 0473-z
44. Lin L, Chu H, Murad MH etal (2018) Empirical comparison of pub-
lication bias tests in meta-analysis. J Gen Intern Med 33:1260–1267.
https:// doi. org/ 10. 1007/ s11606- 018- 4425-7
45. Shen X, Zhao L, Ding G etal (2009) Effect of combined laser
acupuncture on knee osteoarthritis: a pilot study. Lasers Med Sci
24:129–136. https:// doi. org/ 10. 1007/ s10103- 007- 0536-9
46. Helianthi DR, Simadibrata C, Srilestari A etal (2016) Pain reduc-
tion after laser acupuncture treatment in geriatric patients with knee
osteoarthritis: a randomized controlled trial. Acta Med Indones
48:114–121
47. Pinto NC, De MVP, Ferreira NL etal (2022) Customized photo-
biomodulation modulates pain and alters thermography pattern in
patients with knee osteoarthritis: a randomized double-blind pilot
study. Photobiomodul Photomedicine, Laser Surg 40:698–707.
https:// doi. org/ 10. 1089/ photob. 2022. 0067
48. Siriratna P, Ratanasutiranont C, Manissorn T etal (2022) Short-term
efficacy of high-intensity laser therapy in alleviating pain in patients
with knee osteoarthritis: a single-blind randomised controlled trial.
Pain Res Manag. https:// doi. org/ 10. 1155/ 2022/ 13191 65
49. Alqualo-Costa R, Rampazo ÉP, Thome GR etal (2021) Interferen-
tial current and photobiomodulation in knee osteoarthritis: a rand-
omized, placebo-controlled, double-blind clinical trial. Clin Rehabil
35:1413–1427. https:// doi. org/ 10. 1177/ 02692 15521 10120 04
50. Vassão PG, Silva BA, de Souza MC etal (2020) Level of pain,
muscle strength and posture: effects of PBM on an exercise pro-
gram in women with knee osteoarthritis–a randomized controlled
trial. Lasers Med Sci 35:1967–1974. https:// doi. org/ 10. 1007/
s10103- 020- 02989-1
51. Al Rashoud AS, Abboud RJ, Wang W etal (2014) Efficacy of low-
level laser therapy applied at acupuncture points in knee osteoarthri-
tis: a randomised double-blind comparative trial. Physiother (United
Kingdom) 100:242–248. https:// doi. org/ 10. 1016/j. physio. 2013. 09.
007
52. Gworys K, Gasztych J, Puzder A etal (2012) Influence of various
laser therapy methods on knee joint pain and function in patients
with knee osteoarthritis. Ortop Traumatol Rehabil 14:269–277.
https:// doi. org/ 10. 5604/ 15093 492. 10022 57
53 Fukuda VO, Fukuda TY, Guimarães M etal (2011) Short-term effi-
cacy of low-level laser therapy in patients with knee osteoarthritis: a
randomized placebo-controlled double-blind clinical trial. Rev Bras
Ortop. https:// doi. org/ 10. 1016/ s2255- 4971(15) 30407-9
54. Heiskanen V, Hamblin MR (2018) Photobiomodulation: lasers: vs.
light emitting diodes? Photochem Photobiol Sci 17:1003–1017.
https:// doi. org/ 10. 1039/ c8pp0 0176f
55. Kuffler DP (2016) Photobiomodulation in promoting wound healing:
a review. Regen Med 11:107–122. https:// doi. org/ 10. 2217/ rme. 15.
82
56. Brosseau L, Welch V, Wells G etal (2004) Low level laser therapy
(Classes I II and III) for treating osteoarthritis Cochrane database.
Syst Rev. https:// doi. org/ 10. 1002/ 14651 858. CD002 046. pub2
57. Vassão PG, Parisi J, Penha TFC etal (2021) Association of photo-
biomodulation therapy (PBMT) and exercises programs in pain and
functional capacity of patients with knee osteoarthritis (KOA): a sys-
tematic review of randomized trials. Lasers Med Sci 36:1341–1353.
https:// doi. org/ 10. 1007/ s10103- 020- 03223-8
58. Joensen J, Øvsthus K, Reed RK etal (2012) Skin Penetration time-
profiles for continuous 810 nm and superpulsed 904 nm lasers in
a rat model. Photomed Laser Surg 30:688–694. https:// doi. org/ 10.
1089/ pho. 2012. 3306
59. Ferraresi C, Huang Y-Y, Hamblin MR (2016) Photobiomodulation
in human muscle tissue: an advantage in sports performance? J Bio-
photonics 9:1273–1299. https:// doi. org/ 10. 1002/ jbio. 20160 0176
60. Assis L, Moretti AIS, Abrahao TB etal (2012) Low-level laser
therapy (808 nm) reduces inflammatory response and oxidative
stress in rat tibialis anterior muscle after cryolesion. Lasers Surg
Med 44:726–735
61. Pallotta RC, Bjordal JM, Frigo L etal (2012) Infrared (810-nm) low-
level laser therapy on rat experimental knee inflammation. Lasers
Med Sci 27:71–78
62 Riley RD, Jackson D, Salanti G etal (2017) Multivariate and net-
work meta-analysis of multiple outcomes and multiple treatments:
rationale, concepts, and examples. BMJ. https:// doi. org/ 10. 1136/
bmj. j3932
63. Patole S (2021) Network Meta-Analysis BT - Principles and Practice
of Systematic Reviews and Meta-Analysis. In: International S (ed)
Patole S. Publishing, Cham, pp 169–176
64. Fan T, Ruan G, Antony B etal (2021) The interactions between
MRI-detected osteophytes and bone marrow lesions or effusion-
synovitis on knee symptom progression: an exploratory study. Oste-
oarthr Cartil 29:1296–1305. https:// doi. org/ 10. 1016/j. joca. 2021. 06.
008
Publisher's Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com