Efficacy of low-level laser therapy in accelerating tooth movement, preventing relapse and managing acute pain during orthodontic treatment in humans: a systematic review

The search terms used in the problem specification were defined on the basis of the United States National Library of Medicineʼs Medical Subject Headings (MeSH) prior to the literature search.

Three databases were searched to identify all relevant studies: Medline (via PubMed), The Cochrane Controlled Clinical Trials Register and Scitation. The entrez date was 27/11/2015. To ensure the most comprehensive search, no MeSH terms were used, in order to avoid exclusion of recently published studies without these terms. The search strategy is presented in Table 1. The search was assisted by the staff at the Library, Malmö University, Sweden.

Inclusion criteria were determined prior to reading the retrieved abstracts, using the population intervention control outcome method (PICO), as presented in Table 2. Sample size calculations in two studies with sufficient power [7, 8], resulted in the decision to require a minimum of twenty subjects per group.

Publications written in English or a Scandinavian language, addressing questions which seemed relevant to the specifications of the problem, were read in full and either included for further analysis, or excluded. The reference lists of included studies were hand searched for additional publications.

A data extraction protocol (not shown) was used to create an overview of the included studies. See Tables 3 and 4.

The quality of the selected publications was assessed according to predetermined criteria for methodology and performance. The criteria of the checklist for clinical trial of The Swedish Council on Technology Assessment in Health Care (SBU), was modified and used (Appendix 1). Seven variables were analyzed; adequate selection, blinding, adequate interpretation of results, adequate reporting of attrition, adequate reporting of side effects, risks for conflict of interest and adequate study population. Each variable consisted of several subheadings.

The results were summarized and resulted in a yes or a no for the field. One point was awarded for each variable, except for the variable double blinded which was awarded two points, as double blind studies are preferred. The quality of studies awarded six to eight points was denoted as high, three to five points as moderate, and up to two points as low. Quality of evidence was rated according to Grading of Recommendations Assessment, Development and Evaluation (GRADE) guidelines, as strong, moderate, low and very low. To investigate the risk of publication bias, a search was conducted in www.​controlled-trials.​com and www.​clinicaltrials.​gov to verify the number of ongoing studies in this field. No studies were to be found.

The systematic search approach, further described in Fig. 1 yielded three studies [7, 9, 10]. One trial was from India, one from Iran, and one was from Turkey. One study was designed as a randomized controlled trial (RCT) and two as controlled clinical trials (CCT), Table 3. Two of the studies reported a significant increase in velocity of tooth movement [7, 9]. One study showed an increased velocity of tooth movement of approximately 30 % in the laser treatment group compared to the control group [9]. Another study reported similar results after complete canine retraction: acceleration of 27 % in the maxilla and 31 % in the mandible [7]. Finally, one study reported no accelerated tooth movement [10], Table 3.

The systematic search approach failed to identify any relevant study that matched the inclusion criteria, Fig. 2.

The systematic search approach yielded thirteen studies, Fig. 3. Two trials were from Brazil, one from Colombia, one from India, three from Iran, three from Japan, one from Korea, one from Singapore and one was from Spain. Ten studies were designed as RCT and three as CCT, Table 4. Eleven studies showed a statistically significant reduction in reported pain among the patients treated with LLLT [7, 8, 11‐19]. Two studies [10, 20] found no differences in pain sensation, Table 4.

Several studies were excluded because they did not meet the established inclusion criteria. The main reason for exclusion was that the stated objectives did not correspond with the specifications of the research question to be addressed by the review. Four studies were excluded because of inadequate sample size [21‐24]; all but one [22] reported significant acceleration of tooth movement. One study showed that irradiated teeth, compared to control teeth, moved 34 % further during the same time interval and one study showed that LLLT accelerated the initial phase of canine retraction [21, 24].

Although there are few published investigations in this field, all studies included reported similar results, LLLT accelerates tooth movement by 30 %. Doshi-Mehta et al. [7] investigated both velocity and pain. Different exposure times and output power were used to promote analgesia or biostimulation. Given the differences between power and exposure in comparison with other studies, the possibility that the analgesic regimen affected biostimulation and vice versa, cannot be disregarded.

Neither the included nor the excluded studies adequately addressed side effects of LLLT treatment. The clinical advantages and disadvantages must be considered before LLLT becomes generally available for clinical application. Rapid tooth movement increases the risk of root resorption [25], yet only one study [7] used radiographs to monitor possible radiographic changes. It is important to monitor such side effects, even though this is not a primary effect of the irradiation, but an effect of its ability to accelerate tooth movement.

Two studies [7, 9] stated that LLLT reduced orthodontic treatment time: according to the authors this could lead to further benefits for the patient as well as reduced costs. However, another study [10] showed that LLLT did not reduce treatment time. Thus to date the effects of LLLT on treatment time are unconfirmed.

One study [26], excluded because of the small sample size (n = 14), investigated the impact of LLLT on preventing relapse, by stimulating bone remodelling after closure of a median diastema, but there was no statistically significant difference between test and control groups. As no studies were included, the question of whether LLLT can prevent relapse remains undetermined. One reason for the limited number of studies in the field might be the difficulty of study design: an extended follow-up time, preferably up to several years, is desirable. In addition, the long term side effects of using LLLT seems not to be investigated. Nevertheless, given the increasing demand from patients for long-term treatment results, this field of research is likely to warrant more attention in future.

Four studies [23, 27‐29] on LLLT and modulation of pain during orthodontic treatment were excluded due to small sample sizes. All but one [28] of these studies reported reduced pain sensation in the LLLT group. One study [29] showed both less pain and a decrease in Prostaglandin E2 production and two studies [23, 27] showed lower pain prevalence when using LLLT.

Of the included studies [7, 8, 11‐19], all but one [12] had a placebo group. The placebo groups received only light from the laser device or were irradiated with a Light Emitting Diode, LED. Since all studies scored severity of perceived pain by VAS (Visual Analogue Scale) or NRS (Numeric Rate Scale), a placebo group must be considered preferable, as it excludes any response that could interfere with perception of pain. The means used to elicit pain differed in the studies, some using elastomeric separators and others fixed orthodontic appliances. None of the studies addressed the question of whether pain elicited by an elastomeric separator is as recalcitrant as that elicited by a fixed orthodontic appliance. No correlation was discerned between the type of pain stimulus and the study results. Thus, as the method of pain induction seemed to have little impact on the result, it would be more clinically relevant to measure pain associated with fixed appliance treatment rather than separators.

In the studies on pain, the most frequently used method for measurement was VAS. In some studies [11, 17, 19] NRS was used instead. Only one study [7] used a childrenʼs VAS. None of the studies addressed the question of whether the younger participants were able to comprehend the method being applied.

Acute orthodontic pain lasts up to 7 days [30]. It is therefore of interest to note that in the study investigating the severity of pain on day 30 [7], canine retraction did not start until day 21. In one study [19] the subjects rated their pain for 14 days: this must be considered an unnecessarily long follow-up time. Moreover, five subjects in the control group experienced pain until day 14: this is difficult to explain and was not commented on by the authors.

Three studies on pain [11, 17, 19] were double blinded. Blinding was not discussed in any of the studies; although in this context, the risk of operator bias is considered to be low, double blinding would have been preferable. An inherent risk associated with the split mouth method is that the desired effect may occur on the control side as well. This issue was not addressed in any of the studies.

It is notable that none of the studies discussed side effects of laser treatment. Furthermore, no safety instructions appear to have been given to those operating the equipment. LLLT is unlikely to cause side effects in the oral environment, but should always be handled with care [31]. Although eleven out of thirteen studies reported significant modulation of acute orthodontic pain associated with application of LLLT, it was difficult to draw any conclusions because of the variation in study design. Some studies [11, 14] measured the most severe pain as the main outcome, whereas others [15, 19] focused on delayed pain or acute pain. Furthermore, the pain rating was generally low in both the placebo/control group and in the experimental group. A pain reduction of approximately one unit on the scale must unfortunately be considered to be of limited clinical relevance.

Because the studies used different definitions of pain frequency, intensity, onset and duration, these characteristics were not considered separately in the present review. The question arises as to how these aspects of pain perception might affect patient preferences. Would it be preferable to experience severe pain of short duration or mild pain over a longer period? The findings of pain modulation in the studies should be considered in the context of current knowledge about different perceptions of pain. As in all discussions of pain, the wide individual range in sensitivity needs to be taken into account.

Several studies included in this review reported quite promising results for the application of LLLT to accelerate tooth movement and modulate acute pain. In addition, two systematic review were published recently, one meta-analysis on the efficacy of LLLT for accelerating tooth movement and one on LLLT for orthodontic pain, indicating that LLLT might be a promising method to speed up the tooth movement and relieve pain during orthodontic treatment [32, 33]. However, the previous reviews had different inclusion criteria partly identifying other studies compared to the present investigation, which makes it difficult to do any comparisons of the outcome.

In this study, the laser regimens varied widely between the investigations and it is obvious that there is no consensus with respect to different lasers, frequencies and powers. Thus whether the relationship between the different laser parameters is a major determinant of effectiveness of LLLT in improving orthodontic treatment, is still open to speculation. The question of selection of laser regimen cannot be overemphasized. For instance, comparison of studies is difficult because of the confusion of concepts and terms. One example is the term dose, which can be referred to as J/cm² or just Joules. Also, J/cm² can be described as total time or per second. As it is unclear what J/cm² refers to in the included studies, the term dose in Tables 4 and 5 is not further defined.