Hong-Meng Lim BDS, MDS Orthodontics, MORTH RCS (Edin), Kenneth K.K. Lew MDS
Orthodontics, BDS, FDSRCS (Edin) and David K.L. Tay BDS, MS.
Singapore, Available online 31 October 2005.
Abstract : Low level laser therapy (LLLT) has been shown to produce analgesic effects in many clinical applications. The aim of this clinical study was to test the efficacy of LLLT in controlling orthodontic postadjustment pain. Thirty-nine volunteers were selected for this study that used a double-blind design with placebo control. Elastomeric separators were placed at the proximal contacts of one premolar in each quadrant of the dentition to induce orthodontic pain. The tip of a 30 mW gallium-arsenide-aluminium (830 nm) diode laser probe was then placed at the buccal gingiva and directed at the middle third of the root.
Three different treatment durations of 15, 30, and 60 seconds and one placebo treatment of
30 seconds were tested within each subject. The study was conducted over 5 days, and the visual analogue scale (VAS) was used to quantify the pain experienced by the subjects before and after laser applications for each day. Analysis of the VAS median scores showed that teeth exposed to laser treatment had lower levels of pain as compared with those with the placebo treatment. However, nonparametric statistical analysis of the data showed that the differences between treatments and placebo within each subject were not statistically significant. (AM J ORTHOD DENTOFAC ORTHOP 1995;108:614-22.)
INTRODUCTION
It is a well-known clinical observation that after an application of force there will be an initial period of discomfort or pain lasting about 2 to 4 days.
1 , 2 , 3 , 4 and 5 Severe pain was linked to the application of excessive force.
3 , 6 , 7 and 8 According to Proffit, 3 light forces are the key to avoiding pain concomitant with orthodontic treatment. Burstone 6 also noted that the

duration of pain was increased with increased force levels used. It has been postulated by the proponents of the traditional pain-force relationship that, on the basis of the findings of the classical histologic studies, 7 , 9 , 10 and 11 heavier forces would cause greater compression of the periodontal ligament with more tissue damages and, correspondingly, would result in more severe pain responses. However, pain or discomfort was still experienced by most patients even when supposedly physiologic and light forces were used.
2 Clinical studies by several investigators could not find a relationship between pain severity and force level.
1 , 12 , 13 , 14 , 15 and
16 In summary, the relationship between pain and force levels is still controversial. Owing to the wide individual variation, the presence of pain does not mean that there is underlying irreversible tissue damage, whereas tissue damage may not always be accompanied by pain.
Many patients have avoided orthodontic treatment because of the fear of pain, 17 and 18 and it has been proposed that pain from orthodontic treatment could have prevented the patients from achieving effective plaque control.
19 The use of pharmacologic analgesics has its attendant side effects and is contraindicated in patients who are allergic to those drugs. It has been suggested that tooth movement could be affected in patients taking nonsteroidal antiinflammatory drugs (NSAID).
20 As yet, there is no clinically proven, noninvasive, nonpharmacologic method to alleviate pain experienced by the orthodontic patients except for chewing a stick of gum or a plastic wafer immediately after orthodontic activation or adjustment.
3 and 21 However, gum chewing must be instituted before the pain sets in, that is, within the first 8 hours after activation.
3 This method would be ineffective when the teeth become too tender for repetitive chewing, on gum or on a plastic wafer, to be tolerated.
Low level laser has been shown by many investigators to produce analgesic effects in various therapeutic and clinical applications.
22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 and 36 Depending on the energy output and focus, laser can be classified into high, medium, or low power laser.
Terminology for the low power laser included soft laser, mid laser, low energy laser, as well as cold
laser. Low level laser therapy (LLLT) is the new internationally accepted designation 32 and is defined as laser treatment in which the energy output is low enough so as not to cause a rise in the temperature of the treated tissue above 36.5° C or normal body temperature. Because of its lower energy output and intensity, its effects are mainly nonthermal and biostimulatory. The biostimulatory effects of LLLT have been reported by several

investigators.
37 The mechanisms of laser analgesia have not been established, but it has been attributed to its antiinflammatory and neuronal effects.
38 It was proposed by Harris 38 that low level laser irradiation has a benign stimulatory influence on depressed neuronal and lymphocyte respiration. Other neuronal effects include stabilization of membrane potential and release of neurotransmitters.
39 , 40 , 41 and 42 The transmission of laser through tissue is highly wavelength specific and is most optimal in the optical window of 500 to 1200 nm.
38 The 830 nm wavelength of the gallium-arsenide-aluminium (Ga-As-Al) diode laser lies in this optical window and it has been shown to have the greatest tissue penetration when compared with the other laser system.
43
The main objective of this clinical investigation was therefore to determine the clinical efficacy of the low level laser therapy with the Ga-As-AI diode laser as a method to reduce orthodontic postadjustment pain. Specifically, we wanted (1) to determine whether low level laser therapy could provide immediate relief for orthodontic postadjustment pain, (2) to determine whether low level laser therapy could alter the time course of the intensity of orthodontic post-adjustment pain, and (3) if pain relief could be achieved by laser application, to determine the shortest duration of laser application needed to achieve that effect.
Materials and methods
The study employed a within-subject experimental design 44 with each subject being exposed to four treatment parameters (15, 30, and 60 seconds of laser treatment and 30 seconds of placebo application). Within each subject, a multigroup pretest-posttest design with placebo control 45 was used. One premolar in each quadrant of the dentition constituted the basic experimental unit. Two factors were considered separately: (1) the immediate pain relieving effect of the different treatment parameters and (2) the effect of the different treatment parameters on the time course of pain intensity over 5 days. To exclude experimenter's biases and placebo effect in the subjects, the study was conducted double blind. The sequence of application of the four treatment parameters were randomized among the subjects by the Latin Square method.
46 and 47 The purpose of this procedure was to control for the site variability in the different quadrants of the dentition. Any effects that are not due to

the treatment but are due to the inherent differences between treatment sites would be canceled out by this randomization .
Thirty-nine volunteer dental students (aged 21 to 24 years) were chosen for the study. Each subject had to fulfill the following criteria: (l) intact upper and lower dental arches with no open interproximal contacts and at least one premolar in each quadrant of the dentition, and
(2) the teeth must also be free from any acute or chronic pathologic conditions.
The laser unit used was a Class 3B 48 Ga-As-Al diode laser probe by P-laser System (P-Laser
System International, Egedalsvej, Vekso) (Fig. 1) with a wavelength of 830 nm.
Display Full Size version of this image (9K)
Fig. 1. P-Laser system laser probe.
The laser beam emitted in a constant wave with mean output of 30 mW. The intensity of this laser beam was worked out to be 59.7 mW/cm 2 . Other similar low level laser systems used in other investigations include the Panalas-4000 (Japan Medical Laser Laboratory and
Matsushita Electric Co., Tokyo, Japan) and Proton Plus (RONVIG Instruments, Daugaard,
Denmark).
In America, the Food and Drug Administration (FDA) 49 has only approved the use of a dental laser for soft tissue procedures (oral soft tissue surgery) and polymerization of dental restorative materials. With regard to the use of the laser in pain relief, the most frequently used low level lasers in the United States are the infrared (830 nm) gallium-aluminiumarsenide lasers and the visible (633 nm) helium neon lasers.
37 As the use of laser in pain relief is still experimental, informed consent was obtained from all subjects to comply with the
“Guidelines for the use of human subjects in dental research (Council on Dental
Research).” 50 The students were aware that they could withdraw from the study at any time without giving any reason.

The visual analogue scale (VAS) 51 was used in this study to quantify the pain levels (Fig. 2).
Display Full Size version of this image (94K)
Fig. 2. Visual Analogue Scale (VAS) score card used in this study.
The subjects were instructed to mark with an “X” on the scale corresponding to the pain that they were experiencing. The VAS score is defined as the distance in centimetres (up to one decimal place) from the left extreme of the line to the “X” marked by the subjects.
Unitek Alastik TM Quik Stik S-2 separators (Unitek Corp./3M, Monrovia, Calif.) were used to induce orthodontic pain by causing separation of the teeth. The Alastik force module separator plier was used to apply the separators onto the proximal contacts. The separators were placed at the mesial and distal proximal contacts of one premolar in each quadrant of the dentition. Before this, a preseparation VAS score was taken to exclude any pre-existing painful conditions. Five minutes after the placement of separators, the pretreatment VAS scores were recorded. The laser probe was then applied onto the buccal mucosa of the premolar overlying the middle third of the root and the posttreatment scores for each premolar were recorded immediately after the laser exposure. The separators were left in situ for 5 days. On the second, third, fourth, and fifth days, the subjects were again instructed to record the pretreatment and the posttreatment scores. As the intensity of orthodontic postadjustment pain was found to peak around 18 to 36 hours, 4 the duration of the study would cover this period adequately.
In accordance with our objectives, two null hypotheses were tested. The first null hypothesis stated that there was no therapeutic immediate pain relieving effect from the low level laser therapy. The second null hypothesis stated that there was no difference in the time course of pain intensity between the treatment groups and the placebo group. The Friedman two-way analysis of variance of ranks 52 and 53 was used as the main test for statistical inference. It is the

nonparametric equivalent to the parametric one-way analysis of variance (ANOVA) and therefore the two factors were tested separately with the Friedman's test. The procedure of the Friedman's test was described in detail by Siegel 52 and Glantz.
53 The formula for calculating the Friedman's test statistic based on the rank sum for statistical significance is as follows: where R t
denotes sum of ranks for treatment t ∑denotes the sum over all the treatments k is the number of treatments n is the number of subjects
The smaller the value of X 2 r
, the less of a pattern there is relating the ranks within each subject to the treatment. The calculated X 2 r
values were then compared with the values in the table for Chi-square (X 2 r
) distribution.
44
In a pretest-posttest experimental design, one of the appropriate methods of statistical analysis of the data is to take the differences between the pretreatment and posttreatment scores (gain scores) as the basic data rather than to use the absolute VAS scores for comparisons.
45 The computation of the data by the Friedman's test was done by the Statview
II computer statistical analysis program.
Results
The medium VAS scores for each of the four treatment parameters on each day are tabulated in Table I.
Display Full Size version of this image (33K)

The time course of the VAS pain scores over the 5 days is represented by the composite plots of all the median VAS scores over 5 days in Fig. 3.
Display Full Size version of this image (47K)
Fig. 3. Time course of pretreatment and posttreatment VAS median scores for each of four treatment groups recorded over 5 days.
The histogram provides an descriptive trend of the pretreatment and posttreatment median scores of each treatment group for each of the 5 days. It could be seen that the pretreatment scores of the placebo group were the highest among the four groups on the second and third day, whereas the pretreatment scores of the 30- and 60-second treatment groups were the lowest. Another point to note was that the posttreatment scores were generally lower than the pretreatment scores. It must be pointed out that this apparent immediate pain relieving effect was also noted in the placebo group.
The results of the statistical comparisons are shown in Table II and Table III.
Display Full Size version of this image (37K)
Display Full Size version of this image (36K)

Table II shows the rank sums of the differences (gain scores) between the pretreatment and posttreatment scores of each treatment group for each day. These scores allow a comparison of the immediate pain relieving effect of the four treatment parameters. The lower the rank sum, the better the effect. No trend in the rank sums was noted. The computed Friedman statistics (X 2 r
) of the gain scores for each day are also listed at the bottom of the table. The
X 2 r
values were corrected for ties (which occurred when there were two or more equal values), to give more representative X 2 r
values. Comparisons of the X 2 r
values of the gain scores between pretreatment and posttreatment scores for each day, with the critical X 2 r value (X 2 r
crit) at the chance probability level of p = 0.05 with 3 df (X 2 r
crit = 7.815), showed that there was no overall statistically significant differences between the treatment groups and placebo group, except for that between the pretreatment and posttreatment scores of the fourth day (X 2 r
= 8.468). However, analysis of the mean ranks and rank sums of the gain scores for the fourth day showed that the placebo group had the lowest values (mean rank of
2.141 and rank sum of 83.5). The mean ranks and rank sums of the three treatment groups were very close in values. This observation would not be expected if the laser therapy had any immediate pain relieving effect. Therefore the differences noted here were also probably due to chance.
Table III shows the rank sum of the gain scores between the pretreatment scores of each subsequent day and the baseline postseparation pain scores (pretreatment scores of first day).
These allowed comparison of the time course of pain intensity among the four groups. A weak trend was noted on day 2 and day 3 with the 30- and 60-second treatment groups showing lower rank sums. The rank sums of the gain scores between the pretreatment scores of day 2 and day 1 of the 30-and 60-second treatment groups were 90.5 and 96.5, respectively, compared with 98 and 105 for the placebo and 15-second treatment groups, respectively. The rank sums of the gain scores between the pretreatment scores of day 3 and day 1 of the 30- and 60-second treatment groups were both 89.5, compared with 97.5 and
93.5 for the placebo group and 15-second treatment group, respectively. This trend could mean that the increase in pain levels was lesser for the treatment groups of 30 and 60 seconds when compared with the 15-second treatment group and placebo group on day 2 and day 3. However, as in Table II , comparisons of the Friedman's X 2 r
values of the gain

scores for the between day pretreatment scores with the same critical value (p = 0.05, 3 df,
X 2 r
crit = 7.815) showed that there was no statistically significant overall difference between all the gain scores for each of the between-day pretreatment scores comparison.
Discussion
Visual analogue scale as a method of pain measurement has been reviewed extensively and was found to be a reliable method.
5 , 54 , 55 , 56 , 57 , 58 , 59 and 60
The measurements of the marking “X” were made with a millimeter ruler to an accuracy of
±0.05 cm. Any measurement errors within ±0.05 cm would probably not be clinically significant in reflecting the pain responses within each subject. Sampling error could arise in this study because the subjects constituted an experimentally accessible population. Subjects were not selected at random but enrolled on a voluntary basis. In addition, being dental students, they are from a highly defined population group with regard to their dental experiences and attitudes to dental pain and fear. Although the use of the dental students reduced somewhat the emotive and attitudinal variables of pain responses and thus provided a more objective assessment of the treatment, the conclusions drawn from this sample might not fully apply to the general orthodontic patient population. Even among the dental students, the voluntary nature of the enrollment of subjects could have excluded those subjects with lower pain thresholds and different attitudes to dental treatment.
The results of this study were validated by the various features of the experimental design such as within-subject design, 44 double blindness, latin square method of randomisation, 46 and
47 and placebo control. Within-subject design served very well in this study to control for confounding factors. Experiments with such design are very sensitive to how the treatment affects each person. With between-subject design, the changes due to treatment may be masked by the variability in pain responses between subjects. A weakness of the withinsubject design is that it assumed that there was no appreciable between site interaction through the systemic (crossover) effect of the low level laser therapy. Multiple treatment interference through systemic effect in this study could have invalidated the placebo control as it would be under the indirect influence of treatment as well.

The pain scores were highly subjective and should be treated as ordinal data. The statistical fallacy of using the mean to describe ordinal data and its distribution was discussed by Lee”' and Health.
62 Median is preferable to the mean as a measure of central tendency of a sample when the data do not follow a normal distribution or when the measurements are on an ordinal scale. The mean is very sensitive to a skewed distribution, the presence of outliers, and it assumes that the scores are equidistant. Nonequivalence is a general condition in all ordinal subjective data. Since the data from the visual analogue scale (VAS) would be best analyzed on the ordinal scale, a nonparametric version of the repeated measure analysis of variance suitable for within-subject experimental design (Friedman's test) was used.
No trend was noted for the gain scores between pretreatment and posttreatment scores for each day ( Table II ). From the data, it appeared that the low level laser therapy with 15-, 30-, and 60-second exposures could not provide immediate therapeutic pain relief as both descriptive data and statistical tests did not reveal any trend between treatment and placebo.
The apparent immediate pain relieving effect of the treatment as evidenced by the lower posttreatment pain scores as compared with the pretreatment pain scores ( Table I , Fig. 3 ) was most probably due to the Hawthorne's effect.
45 Hawthorne's effect refers to two phenomena: (1) The subjects' knowledge that they were participating in a clinical trial of the analgesic efficacy of the low level laser could have induced them to report lower pain scores after laser application. This is known as the perceived demand characteristics of the experiment by the subjects. (2) There could be the true placebo effect due to the genuine belief that the treatment had an effect.
From the gain scores of the pretreatment scores of between-day comparisons ( Table III ), there was a weak trend of lower ranks for the 30- and 60-second treatment groups in the day
2-day 1 and day 3-day 1 comparisons. Although the trend was only noted on the second and third day and the differences were small and not statistically significant (at p < 0.05), it could be that the treatment effects were small and only managed to show a weak trend on the second and third day when orthodontic pain normally peaked in intensity.
4 and 5 These smaller gain scores for the 30- and 60-second treatment groups reflected lower pain intensities for these two treatment groups on the second and third day. Thus, it was possible that one or

two applications of the low level laser (Ga-As-Al) could have some therapeutic effects on the time course of the intensity of orthodontic pain.
From clinical observations, the placement of separators is known to cause considerable pain although there is much individual variation with some patients reporting no pain at all. The method of orthodontic pain simulation using separators could have produced highly variable pain responses and pain with a magnitude that was not sufficiently high enough to allow the treatment effect to be felt. This could have resulted in the nonsignificant findings. However,
Roth and Thrash 4 used the same method of pain induction to test the efficacy of transcutaneous electrical nerve stimulation (TENS) as a method of pain control and found that TENS could significantly provide relief of orthodontic pain. A point to note is that the authors used parametric analysis of the VAS data, which should have been treated as ordinal data.
One of the problems in the planning of this study was the determination of the size of the treatment effect that we would like to detect and consider clinically significant. This was because pain responses are highly subjective and variable. It must be noted that the lack of statistical significance did not equate that there was no clinical effect. We should be looking more at the general trend of the pain scores rather than the statistical significance of the differences in determining whether there is a potential clinical application for low level laser therapy in orthodontics.
Orthodontic pain has been related to the acute inflammatory responses of the periodontal ligament to orthodontic forces.
63 and 64 Kess et al.
65 showed that prostaglandins level increased and peaked at 24 hours after the application of orthodontic forces, whereas Kamogashira et al.
66 showed that the level of substance P increased and peaked at 36 hours after separation of incisors by orthodontic forces. Prostaglandins has been suggested by Ferreira 67 and Higg 68 to sensitize pain receptors resulting in increased pain sensitivity to hyperalgesic inflammatory mediators, such as histamine or bradykinin. The mechanisms of laser analgesia have been suggested to be due to the direct effect of the laser on the nerve fibers, by stabilizing its depolarizing potential, or to its effects on the cellular and biochemical processes of the inflammatory responses.
38 As no immediate therapeutic pain relief was noted in this study and the treatment effect took about 24 to 48 hours to become apparent, the results of this

study tended to support the hypothesis that laser analgesia was due mainly to the effect of laser treatment on the inflammatory processes.
Laser analgesia is a new treatment modality and has the advantages of being noninvasive, easy to administer, and having no known adverse tissue reactions. It is worthwhile to look into its potential applications in orthodontics. The apparent lack of significant treatment effect in this study could be due to several reasons. First, it could be that the low level laser therapy did not actually have any therapeutic effect. Second, it could be because the treatment effect was too small and was masked by the placebo effect. Finally, it could be due to the multiple treatment interference through systemic effect that invalidated the placebo control. Therefore future investigations should look into increasing the size of the experimental effect by changing the treatment parameters, increasing the magnitude of pain induction, and providing a no treatment control group. The laser energy in joules (J) delivered to the target tissue was worked out to be 0.45 J, 0.95 J, and 1.8 J for the 15, 30, and
60 seconds beam durations, respectively. These dosages were well within the range of 0.5 to
10 J per treatment point recommended by Kert and Rose.
32 Thus there may be room for increasing the dosage per treatment point by increasing the treatment duration.
Conclusions
1. A weak trend of lower pain scores and lower increases in pain intensity in the 30- and 60second treatment groups, on the second and third day, was observed when compared with the placebo group.
2. However, the two null hypotheses of no treatment effect for the two factors considered in this study could not be rejected statistically.
3. Although low level laser was found to be unable to provide immediate pain relief, it could have a potential in reducing the intensity of pain during its time course with just one or two applications. Further investigations into the potential of low level laser therapy, using different treatment protocols and treatment parameters in controlled double blind clinical trials, are warranted.

References
1 JE Soltis, PR Nakfoor and DC Bowman, Changes in ability of patients to differentiate intensity of forces applied to maxillary central incisors during orthodontic treatment, J Dent
Res 50 (1971), pp. 590–596. View Record in Scopus | Cited By in Scopus
2 ML Jones, An investigation into the initial discomfort caused by placement of an archwire,
Eur J Orthod 6 (1984), pp. 48–54. View Record in Scopus | Cited By in Scopus
3 WR Proffit, The biologic basis of orthodontic therapy. In: WR Proffit and HW Fields Jr,
Editors, Contemporary orthodontics, : CV Mosby, St Louis (1986), p. 241.
4 PM Roth, Thrash WJ. Effect of transcutaneous electrical nerve stimulation for controlling pain associated with orthodontic tooth movement, Am J Orthod Dentofac Orthop 90 (1986), pp.
132–138. Abstract | PDF (2172 K) | View Record in Scopus | Cited By in Scopus
5 P Ngan, B Kess and S Wilson, Perception of discomfort by patients undergoing orthodontic treatment, Am J Orthod 96 (1989), pp. 47–53. Abstract | Full Text + Links |
PDF (793 K) | View Record in Scopus | Cited By in Scopus
6 CJ Burstone, Biomechanics of tooth movement. In: BS Krau, Editor, Vistas in orthodontics, :
Lea & Febiger, Philadelphia (1962), pp. 210–213.
7 K Reitan, Selecting forces in orthodontics, Eur Orthod Soc Trans 32 (1956), pp. 108–126.
8 E Storey and R Smith, Forces in orthodontics and its relation to tooth movement, Aust J
Dent 41 (1952), pp. 11–18.
9 G Sandstedt, Eine Betrage zur Theorir der Zanregulierung Nordisk Tandlakare, Tidsskrift 4
(1904), pp. 1–2 1905.
10 AM Schwarz, Tissue changes incidental to orthodontic tooth movement, Int J Orthod 18
(1932), pp. 331–352.

11 K Reitan, Tissue behavior during orthodontic tooth movement, Am J Orthod 46 (1960), pp. 881–900.
12 EH Hixon, H Aitikian, G Callow, H McDonald and RJ Tracy, Optimal force, differential force and anchorage, Am J Orthod 55 (1969), pp. 437–457. View Record in Scopus | Cited By in Scopus
13 AA Gianelly and HM Goldman, Biologic basis of Orthodontics, : Lea & Febiger,
Philadelphia (1971), pp. 164–165.
14 IE Johnson and CH Boester, A clinical investigation of the concepts of differential and optimal force in canine retraction, Angle Orthod 44 (1974), pp. 113–119.
15 GF Andreasen and D Zwanziger, A clinical evaluation of the differential force concept as applied to the edgewise bracket, Am J Orthod 78 (1980), pp. 25–40. View Record in Scopus |
Cited By in Scopus
16 ML Jones and S Richmond, Initial tooth movement: force application and pain—a relationship?, Am J Orthod 88 (1985), pp. 111–116. View Record in Scopus | Cited By in
Scopus
17 BH Tayer and MJ Burek, A survey of adults' attitudes towards orthodontic therapy, Am J
Orthod 79 (1981), pp. 305–315. View Record in Scopus | Cited By in Scopus
18 RG Oliver and YM Knapman, Attitudes to orthodontic treatment, Br J Orthod 12 (1985), pp. 179–188. View Record in Scopus | Cited By in Scopus
19 LW White, Tooth brushing pressures of orthodontic patients, Am J Orthod 83 (1983) (2), pp. 109–113. Abstract | PDF (494 K) | View Record in Scopus | Cited By in Scopus
20 AB Chumbley and OC Tuncay, The effect of indomethacin (an aspirin-like drug) on the rate of orthodontic tooth movement, Am J Orthod 89 (1986), pp. 312–314. View Record in
Scopus | Cited By in Scopus

21 LW White, Pain and cooperation on orthodontic treatment, J Clin Orthod 18 (1984), pp.
572–575. View Record in Scopus | Cited By in Scopus
22 FMW Plog, Biophysical application of the laser beam. In: HK Koebner, Editor, Lasers in
medicine, : John Wiley, Chicester (1980), pp. 21–37.
23 M Kroetlinger, On the use of the laser in acupuncture, Acupuncture Electro-Ther Res, Int J 5
(1980), pp. 297–311. View Record in Scopus | Cited By in Scopus
24 JJ Bischko, Use of the laser beam in acupuncture, Acupuncture Electro-Ther Res, Int J 5
(1980), pp. 29–40. View Record in Scopus | Cited By in Scopus
25 G Calderhead, T. Ohshiro, E Itoh, T Okada and Y Kato, The NdYAG and GaAIAs
Lasers: a comparative analysis in pain therapy, Laser acupuncture 21 (1982), pp. 1–4.
26 J Walker, Relief from chronic pain by low power laser irradiation, Neuroscience Letters 43
(1983), pp. 339–344. Abstract | Abstract + References | PDF (367 K) | View Record in
Scopus | Cited By in Scopus
27 T Kreczi and D Klingler, A comparison of laser acupuncture versus placebo in radicular and pseudoradicular pain syndromes as recorded by subjective responses of patients,
Acupuncture Electro-Ther Res, Int J 5 (1986), pp. 297–311.
28 T Floter, Pain treatment with laser: a double blind study, Acupuncture Electro-Ther Res, Int J
13 (1988), pp. 236–239.
29 O Airaksinen, P Rantanen and P Kolari, Effects of the infrared laser therapy at treated and non-treated trigger points, Acupuncture Electro-Ther Res, Int J 14 (1989), pp. 9–14. View
Record in Scopus | Cited By in Scopus
30 NJ Bezuur, TL Hansson and LLMH Habets, The effect of therapeutic laser treatment in patients with craniomandibular disorders, J Craniomand Disorders 2 (1988), pp. 83–86. View
Record in Scopus | Cited By in Scopus

31 TL Hansson, Infrared laser in the treatment of craniomandibular disorders, anthrogenous pain, J Prosthet Dent 61 (1989), pp. 614–617. Abstract | Full Text + Links | PDF (1596 K)
| View Record in Scopus | Cited By in Scopus
32 J Kert and L Rose, Clinical laser therapy—low level laser therapy, Denmark: Scandinavian
Medical Laser Technology, Ballerup (1989).
33 M Harazaki, Y Isshikii and K Nojima et al., A survey on the pain relief effect following the application of soft laser in orthodontic surgical patients (Abstract), Laser Therapy—An Int
J Low Level Laser Therapy and Photobioactivation 2 (1990) (1), p. 45.
34 C Clokie, KC Bentley and TW Head, The effects of the helium-neon laser on postsurgical discomfort: a pilot study, J Can Dent Assoc 57(1991) (7), pp. 584–586. View Record in Scopus
| Cited By in Scopus
35 K Iijima, N Shimoyama, M Shimoyama and T Mizuguchi, Evaluation of analgesic effect of low-power He-Ne laser on postherpetic neuralgia using VAS and modified McGill pain questionnaire, J Clin Laser Med Surg Apr (1991), pp. 121–126. View Record in Scopus | Cited
By in Scopus
36 M Midda and P Renton-Harper, Lasers in dentistry, Br Dent J 170 (1991), pp. 343–346.
View Record in Scopus | Cited By in Scopus
37 JR Basford, Low energy laser treatment of pain and wounds: hype, hope or hokum
(Editorial), Mayo Clinic Proc 61 (1986), pp. 671–675. View Record in Scopus | Cited By in
Scopus
38 DM Harris, Biomolecular mechanisms of laser biostimulation, J Clin Laser Med Surg Aug
(1991), pp. 277–280.
39 RL Fork, Laser stimulation of nerve cells in Aplysia, Science 171 (1971), pp. 901–908.
40 ES Vizi, E Mester, S Tisza and A Mester, Acetycholine releasing effect of laser irradiation on Auerbach's plexus in guinea pig ileum, J Neural Transm 40 (1977), pp. 305–308. Full Text via CrossRef | View Record in Scopus | Cited By in Scopus

41 JB Walker and LK Akhanjee, Laser induced somatosensory evoked potentials: evidence of photosensitivity in peripheral nerves, Brain Res 344 (1985), pp. 281–285. Abstract | PDF
(380 K) | View Record in Scopus | Cited By in Scopus
42 RN Ponnudurai, VK Zbuzek and WH Wu, Hypoalgesia effect of laser photostimulation shown by rat tail flick test, Acupuncture Electro-Ther Res, Int J 12 (1987), pp. 93–100. View
Record in Scopus | Cited By in Scopus
43 T Ohshiro and RG Calderhead, Development of low reactive level laser therapy and its present status, J Clin Laser Med Surg Aug (1991), pp. 267–275. View Record in Scopus |
Cited By in Scopus
44 RJ Shavelson, Statistical reasoning for the behavioural sciences (2nd ed), : Allyn and
Bacon, Inc, Boston (1988), p. 463.
45 SW Huck, WH Cormier and WG Bounds, Reading Statistics and Research, : Harper and
Row Publishers, New York (1975).
46 M Linton and PS Gallo, The practical statistician: simplified handbook of statistics, :
Wadsworth Publishing Co, Pacific Grove, California (1975), p. 222.
47 P Armitage and G Berry, Statistical Methods in Medical Research, : Blackwell Scientific
Publications Ltd, Oxford (1988), p. 239.
48 Laser classification in radiation safety of laser products (BS7192), : British Standards Institute,
Milton Keynes (1989).
49 Fact sheet: laser biostimulation, : Division of Consumer Affairs, Center for Devices and
Radiological Health, FDA, Washington, DC (1984).
50 ML Darby and DM Bowen, Legal and ethical concerns in oral health research, Research
methods for the oral health professionals—an introduction, : CV Mosby, St Louis (1980), pp. 45–54.
51 In: R Melzack, Editor, Pain measurement and assessment, : Raven Press, New York (1983).

52 S Siegel and NJ Castellan, Nonparametric statistics for the behavioural sciences, :
McGraw-Hill, New York (1988).
53 SA Glantz, Primer of biostatistics (2nd ed), : McGraw-Hill, New York (1988).
54 J Scott and EC Huskisson, Accuracy of subjective measurements made with or without previous scores: an important source of error in serial measurement of subjective states, Ann
Rheum Dis 38 (1979), pp. 558–559. View Record in Scopus | Cited By in Scopus
55 CRB Joyce, DW Zutshi, V Hrubes and RM Mason, Comparison of fixed interval and visual analogue scales for rating chronic pain, Eur J Clin Pharm 8 (1975), pp. 415–420. Full
Text via CrossRef | View Record in Scopus | Cited By in Scopus
56 J Scott and EC Huskisson, Graphic representation of pain, Pain 2 (1976), pp. 175–184.
Abstract | PDF (712 K) | View Record in Scopus | Cited By in Scopus
57 WW Downie, PA Leatham, VM Rhind, V Wright, JA Branco and JA Anderson, Studies with pain rating scales, Ann Rheum Dis 37 (1978), pp. 378–381. View Record in Scopus |
Cited By in Scopus
58 EC Huskisson, Measurement of pain, Lancet 2 (1974), pp. 1127–1131. Abstract | Full
Text + Links | PDF (639 K) | View Record in Scopus | Cited By in Scopus
59 EC Huskisson, Visual analogue scale. In: R Melzach, Editor, Pain measurement and
assessment, : Raven Press, New York (1983), p. 33.
60 DD Price, PA McGrath, A Rafii and B Buckingham, The validation of visual analogue scales as ratio scale measures for chronic and experimental pain, Pain 17 (1983), pp. 45–56.
Abstract | Abstract + References | PDF (952 K) | View Record in Scopus | Cited By in
Scopus
J Lee, Statistical analysis of data that are ordinal and highly subjective, J Singapore Nat Acad
Science 14 (1985), pp. 98–100.

62 DF Health, Morbidity and mortality of car occupants: uses of the injury severity score, Br
Med J 290 (1985), p. 319.
63 L Furstman and S Bernick, Clinical considerations of the periodontium, Am J Orthod 61
(1972), pp. 138–155. View Record in Scopus | Cited By in Scopus
64 K Yamasaki, Y Shibata, Y Shibata and T Fukuhara, The nature of pain reaction associated with orthodontic tooth movement, J Jpn Orthod Soc 44 (1985), pp. 332–338. View Record in
Scopus | Cited By in Scopus
65 B. Kess, B Zwilling, R Lanese, J Shanfeld and Z Davidovitch, The effect of indomethacin on periodontal PGE content during orthodontic treatment, J Dent Res 66 (1987), p. 332.
66 K Kamogashira, M Yanabu and K Ichikawa et al., The effects of upper incisor separation on the submandibular and sublingual glands of rats, J Dent Res 67 (1988), pp. 602–610. View
Record in Scopus | Cited By in Scopus
67 SH Ferreira, M Nakamura and M Castro, The hyperalgesic effects of prostacyclin and prostaglandin E
2
, Prostaglandins 16 (1978), pp. 31–37. View Record in Scopus | Cited By in
Scopus
68 GA Higgs and S Moncada, Interactions of arachidonate products with other pain mediators. In: JJ Bonica, U Lindbolm and A Iggo, Editors, Advances in pain research and therapy
5, : Raven Press, New York (1983), pp. 617–626.
American Journal of Orthodontics and Dentofacial Orthopedics
Volume 108, Issue 6 , December 1995, Pages 614-622