According to our search in the databases, this is the first systematic review and meta-analysis conducted to evaluate the effect of adjunctive use of PBM in periodontal surgical procedures leaving a secondary intention healing wound, such as gingivectomy, harvesting grafts from donor sites or depigmentation procedures, to assess its effectiveness on gingival healing and pain relief. All of the included studies were clinical trials, and the included non-randomized trials satisfied the pre-quality assessment.
Several studies have stated that PBM can facilitate the speed and quality of wound healing through different mechanisms. Keskiner et al. reported an increase in palatal wound fluid (PWF), transforming growth factor-b1 (TGF-b1), platelet-derived growth factor-BB (PDGF-BB), and interleukin-8 (IL-8) levels. This might indicate an increased rate of wound healing by stimulation of the secretion of selected mediators . Enhanced collagen production and increase the levels of growth factors and extracellular matrix-remodeling proteins, and stimulation of synthesis of adenosine triphosphate, fibroblastic proliferation, and angiogenesis in a dose-dependent manner have also been reported [16, 17, 19, 20]. Moreover, we can assume that improved pain relief, improved re-epithelialization, and tissue thickness could be the direct impact of the improved wound healing process.
In a recent meta-analysis, it was stated that the mechanism and effect of PBM on primary or secondary wound healing might be significantly different . This might be due to different healing mechanisms and cellular and molecular events between secondary and primary wound healing. Secondary healing involves more granulation and collagenous tissue formation in the proliferation stage and a higher amount of remodeling and contraction in the remodeling stage of wound healing. Moreover, secondary healing is associated with more tendency to wound infection and leaves more scar tissue in the wound site . Therefore the application of PBM may more beneficial in these patients. In this study, we have only assessed the effect of PBM on secondary intention healing gingival wounds.
Although the included studies in the present review have all used laser wavelengths in the red and near infra-red range; they showed a great variation in irradiation parameters and the method of application of the adjunctive PBM therapy, making it challenging to draw evidence-based conclusions regarding the most appropriate irradiation settings needed for improvement in healing and pain relief. The most suitable laser settings for biostimulation of healing and reduction of post-operative pain of periodontal surgical wounds have not been determined yet due to the great variation observed in irradiation parameters in the available literature. Factors, such as the diameter of the fiber, can alter power density and energy output in the application of lasers. It could also change the quantity of energy that is applied during the treatment, altering the wound-healing effect of PBM.
The included studies had utilized various wavelengths and irradiation parameters for PBM of the wounds. Only one study had used an LED 650nm device and in one study PBM was performed using an Nd:YAG laser (1084nm) device. The other included studies had used diode lasers with red to near infra-red wavelengths (588-970nm) for irradiation of the surgical sites. Energy densities ranged from 1.6 -8.6J/cm2. The majority of studies had applied an energy density of 4J/cm2 per point. However, the output powers ranged from as low as 0.05 W to 5 W.
In the present study, despite all methodological variations, the results of the meta-analysis of the Landry wound healing index and complete wound epithelialization demonstrated a statistically significant improvement in secondary wound healing after periodontal surgeries. One of the studies from which the meta-analysis of the Landry wound healing index was conducted, had a very high quality with no risk of bias . The other study did not blind the outcome assessor and had detection bias with lower quality . Four studies used for the meta-analysis of the complete wound epithelialization and one of them had high quality with low risk of bias . one had incomplete outcome data (attrition bias) , one did not blind the participants or personnel (performance bias) and also had some selective reporting (reporting bias) , and one had issues with randomization process (selection bias) and did not blind the participants or personnel (performance bias) with relatively low quality .
Therefore, it appears that PBM can be beneficial in improving secondary wound healing after certain types of periodontal surgeries. However, the included studies showed some controversies about the efficiency of PBM on post-operative pain. These results may be due to two factors: Firstly, pain measurement is subjective. Secondly, although the VAS scale is a valid method, the range of results is widely heterogeneous . Moreover, the method of pain sensation evaluation varied in the studies. For example, some of the studies used external stimuli to measure pain. However, a recent systematic review on photobiomodulation and acute pain has indicated positive results for PBM and reported similar effects to NSAIDs consumption .
The summary of the risk of bias evaluation is shown in Figure. 2. The main source of bias in the included studies was the performance bias, which shows that most of the studies did not focus on blinding the participants and personnel. To reduce this type of bias in the studies, the conductors could use sham lasers in the control sites. Or they could use the same laser in the control sites without pressing the button just to mimic the application of PBM. Moreover, To blind the operator; a person not involved in the study design could be asked to activate the laser in the specified sites.
Another main source of bias was the selection bias. Random sequence and concealment of allocation are considered of great importance in any study. Future studies should pay more attention to these risks in their studies.
In the present study, we did not include studies in which the patients had specific risk factors such as smoking or diabetes as there was not enough data on possible systemic complications and PBM therapy. Also, due to incomplete information and methodological heterogeneity, variable laser parameters, and methods of its application, the authors couldn't perform a meta-analysis for all of the variables in the included studies. Regarding post-operative pain and discomfort, considerable heterogeneity existed among the evaluation methods. For instance, in some studies, VAS was evaluated by application of an external stimuli like air spray [46, 48], while in other studies no stimulation method was utilized. One of the included studies used the NRS index to evaluate post-operative pain . Overall, because of the different methods used to evaluate post-operative pain in these studies, the criteria for a meta-analysis were not met.
Furthermore, when assessing the tissue epithelialization, the existing diversity in the used methods like the evaluation of pictures taken from intervention areas , visual inspection of the wound [17, 20, 30, 37, 38], or the use of computer software  did not allow us to perform a meta-analysis. The exact area of the initial wounds was not mentioned in any of the studies, which may be an interesting factor to consider in future study designs evaluating the effect of PBM in wound healing.